Health and Prisoner Reentry: How Physical, Mental, and Substance Abuse Conditions Shape the Process of Reintegration, Urban Institute, 2008

Disciplinary Self-Help Litigation Manual - Header

Health and Prisoner Reentry: How Physical, Mental, and Substance Abuse Conditions Shape the Process of Reintegration, Urban Institute, 2008

• Feb. 19, 2016 • Topics: Medical, Prisoner Privileges, Release and Reentry

Download original document:
Brief thumbnail

Document text

This text is machine-read, and may contain errors. Check the original document to verify accuracy.

R E S E A R C H

research for safer communities

February 2008

Kamala Mallik-Kane
Christy A. Visher

R E P O R T

Health and Prisoner
Reentry: How Physical,
Mental, and Substance
Abuse Conditions Shape the
Process of Reintegration

URBAN INSTITUTE

Justice Policy Center

URBAN INSTITUTE
Justice Policy Center
2100 M Street NW
Washington, DC 20037
www.urban.org

© February 2008. The Urban Institute.
The views expressed in this report are those of the authors, and do not necessarily
represent the official position or policies of the Urban Institute, its trustees, or its funders.

Health and Prisoner Reentry:
How Physical, Mental, and Substance Abuse
Conditions Shape the Process of
Reintegration

Kamala Mallik-Kane
Christy A. Visher

About the authors

Kamala Mallik-Kane, MPH, is a research associate in the Justice Policy Center at the
Urban Institute, where her work focuses on criminal justice and public health issues. Ms.
Mallik-Kane has worked extensively on prisoner reentry issues over the past four years, serving
as an analyst on the multistate Returning Home project. She is also a member of a research team
evaluating criminal justice interventions for mentally-ill persons. Before joining the Urban
Institute, Ms. Mallik-Kane worked as a liaison to the Health Resources and Services
Administration and as an epidemiologist for the Baltimore City Health Department.
Christy A. Visher, Ph.D., is a principal research associate in the Justice Policy Center at the
Urban Institute. She is principal investigator of the multistate Returning Home project, which is
examining the experiences of men and women immediately prior to and following their release
from prison, as well as up to a year into their reintegration into the community. She is also
coprincipal investigator of a national evaluation of prisoner reentry initiatives.

Acknowledgments

We would like to thank the many individuals and organizations that have contributed to the
success of this research project. The Returning Home study in Texas and Ohio was funded by the
generous support of the Annie E. Casey Foundation, the JEHT Foundation, the Rockefeller
Foundation, the Robert Wood Johnson Foundation, the Smith Richardson Foundation, the Health
Foundation of Greater Cincinnati, the George Gund Foundation, the Cleveland Foundation, and
the Houston Endowment. The Ohio Department of Rehabilitation and Correction and the Texas
Department of Criminal Justice were invaluable in facilitating access to their correctional
facilities for interviews. We are particularly appreciative of the staff at Research Support
Services, based in Chicago, Illinois, and NuStats, based in Austin, Texas, who conducted the
original data collection for this report. We also thank the Urban Institute’s Justice Policy Center
staff who assisted in the production of this report, including Lisa Brooks and Jennifer Yahner,
who helped prepare the data for this report, Samantha Hetrick and Sara Debus, for their review
of earlier drafts of this document, and Noelle Irving and Aaron Morrissey for their help with
formatting.

Contents

EXECUTIVE SUMMARY ..................................................................................................................... 1
Key findings..........................................................................................................................................1
Policy Implications ................................................................................................................................3

INTRODUCTION ........................................................................................................... 7

PRISONER REENTRY: AN OVERVIEW....................................................................... 9
Profile of returning prisoners ................................................................................................................9
Health status of returning prisoners......................................................................................................9
The “average” reentry experience ......................................................................................................14
Housing ................................................................................................................................................................15
Employment..........................................................................................................................................................16
Family support ......................................................................................................................................................17
Postrelease substance use ..................................................................................................................................18
Criminal involvement ............................................................................................................................................18

PHYSICAL HEALTH AND REENTRY......................................................................... 21
Physical health status and treatment..................................................................................................21
Reentry experiences of prisoners with physical health conditions......................................................24
Housing and physical health.................................................................................................................................24
Employment and physical health..........................................................................................................................25
Family support and physical health ......................................................................................................................27
Substance use and physical health ......................................................................................................................27
Criminal involvement and physical health ............................................................................................................29

Conclusions about physical health and reentry ..................................................................................30

MENTAL HEALTH AND REENTRY ............................................................................ 33
Mental health status and treatment ....................................................................................................33
Reentry experiences of prisoners with mental health conditions ........................................................36
Housing and mental health...................................................................................................................................36
Employment and mental health ............................................................................................................................38
Family support and mental health ........................................................................................................................39
Substance use and mental health ........................................................................................................................40
Criminal involvement and mental health...............................................................................................................42

Conclusions about mental health and reentry ....................................................................................43

SUBSTANCE ABUSE AND REENTRY....................................................................... 45
Substance abuse status and treatment ..............................................................................................45
Reentry experiences of prisoners with substance abuse problems....................................................48
Housing and substance abuse .............................................................................................................................48
Employment and substance abuse ......................................................................................................................50
Family support and substance abuse...................................................................................................................52
Criminal involvement and substance abuse .........................................................................................................53

Conclusions about substance abuse and reentry...............................................................................54

DISCUSSION AND POLICY IMPLICATIONS ............................................................. 57

METHODOLOGY .............................................................................................................................. 63
REFERENCES .................................................................................................................................. 67

Tables and Figures

Table 1. Selected characteristics of Returning Home respondents..............................................................................10
Figure 1. Interplay of physical health, mental health, and substance abuse conditions among returning
prisoners, by gender ....................................................................................................................................................12
Figure 2. Prevalence of physical health, mental health, and substance abuse conditions among returning
prisoners, by gender ....................................................................................................................................................13
Figure 3. Health insurance coverage among returning prisoners, by gender...............................................................13
Table 2. Key reentry outcomes eight to ten months after prison release, by gender ...................................................14
Table 3. Physical health conditions among returning prisoners ...................................................................................22
Figure 4. Percentage of respondents with selected physical health conditions who received treatment, by
gender..........................................................................................................................................................................23
Figure 5. Percentage of respondents who were ever employed, by physical health status and gender ......................26
Figure 6. Sources of income eight to ten months after prison release, by physical health status and gender .............26
Figure 7. Percentage of respondents reporting substance use, by physical health status and gender .......................28
Figure 8. Percentage who participated in substance abuse treatment services, including Alcoholics or
Narcotics Anonymous, by physical health status and gender ......................................................................................28
Figure 9. Postrelease criminal involvement and criminal justice system response, by physical health status and
gender..........................................................................................................................................................................29
Table 4. Prevalence estimates of mental illness among returning prisoners................................................................33
Table 5. Percentage of respondents with mental illness who received mental health treatment, by gender................35
Figure 10. Percentage of returning prisoners who ever lived with family members, by mental health status and
gender..........................................................................................................................................................................36
Figure 11. Rates of homelessness, by mental health status and gender .....................................................................37
Figure 12. Percentage of respondents who were ever employed, by mental health status and gender.......................38
Figure 13. Sources of income eight to ten months after prison release, by mental health status and gender..............39
Figure 14. Percentage who participated in substance abuse treatment services, including Alcoholics or
Narcotics Anonymous, by mental health status and gender ........................................................................................41
Figure 15. Postrelease criminal activity and justice system response, by mental health status and gender ................42
Figure 16. Substance use in the six months prior to incarceration, by gender .............................................................45
Figure 17. Percentage who participated in substance abuse treatment services, including Alcoholics or
Narcotics Anonymous, by substance abuse status and gender ...................................................................................46
Figure 18. Percentage reporting any drug use or alcohol intoxication, by substance abuse status and gender ..........47
Figure 19. Percentage of returning prisoners who ever lived with family members, by substance abuse status
and gender ...................................................................................................................................................................49
Figure 20. Rates of homelessness, by substance abuse status and gender ...............................................................49
Figure 21. Percentage of respondents who were ever employed, by substance abuse status and gender .................51
Figure 22. Percentage of respondents reporting current employment at the time of the interview, by substance
abuse status and gender..............................................................................................................................................51
Figure 23. Financial support from family members, by substance abuse status and gender .......................................52

Figure 24. Postrelease criminal activity and justice system response, by substance abuse status and gender ..........54
Figure 25. Treatment rates for physical, mental, and substance abuse conditions over time compared to
respondent assessments of their overall health, by gender .........................................................................................58
Table 6. Key reentry outcomes, by gender, with significant differences by health condition highlighted......................59

Executive Summary

Each year, nearly 700,000 men and women are released from prison into communities across the
United States but many do not make a successful transition: two-thirds are arrested within three
years and one-half are returned to prison, either for parole violations or new crimes. The Urban
Institute’s study, Returning Home: Understanding the Challenges of Prisoner Reentry, provides
an in-depth examination of the reentry process through a series of interviews with a
representative sample of 1,100 returning prisoners before and after their release. This report
documents the health challenges facing returning prisoners and describes how individuals with
health conditions navigated the first year after release from prison. Taking a comprehensive
perspective on “health,” we report on the influence of physical health conditions, mental illness,
and substance abuse on the reentry process. We demonstrate, empirically, how returning
prisoners with these health conditions faced distinct challenges with regard to finding housing
and employment, reconnecting with family members, abstaining from substance use and crime,
and avoiding a return to prison.
KEY FINDINGS

Nearly all returning prisoners—8 in 10 men and 9 in 10 women—had chronic health
conditions requiring treatment or management.
− One-half of men and two-thirds of women had been diagnosed with chronic physical
health conditions such as asthma, diabetes, hepatitis, or HIV/AIDS.
− Fifteen percent of men and over one-third of women reported having been diagnosed
with depression or another mental illness; the actual prevalence of mental health
conditions is likely to be double the self-reported amount.
− About two-thirds of men and women reported active substance abuse in the six
months before this incarceration.

Returning prisoners often had more than one type of health problem. Roughly 4 in 10 men
and 6 in 10 women reported a combination of physical health, mental health, and substance
abuse conditions, including an estimated one-tenth of men and one-quarter of women with
co-occurring substance abuse and mental health conditions.

Many prisoners with health conditions did not receive treatment while incarcerated, and
treatment rates decreased further upon release.
− Two-thirds of men and three-quarters of women with physical health conditions
received treatment during prison. Eight to ten months after release the shares receiving
treatment for these conditions fell to one-half of men and 6 in 10 women.
− About 6 in 10 men and women with mental health conditions received mental health
treatment in prison. Eight to ten months after release the share receiving treatment
declined to one-half of men and 4 in 10 women.

Health and Prisoner Reentry

−

One-half of men and 4 in 10 women with substance abuse problems participated in
treatment services during prison, including Alcoholics Anonymous (AA), Narcotics
Anonymous (NA), and formal treatment programs. Eight to ten months after release,
one-quarter were still receiving such services.
Individuals who were taking prescription medicines in prison usually received a
supply of medication upon release but, eight to ten months later, about two-thirds
were still using the same type of medicine.

Most returning prisoners (68 percent of men and 58 percent of women) were without
health insurance 8 to 10 months after release. Nevertheless, returning prisoners with
physical and mental health conditions were heavy consumers of health services, including
emergency room visits and hospitalizations.
− Over 7 in 10 individuals with physical and mental health conditions utilized some
health care within 8 to 10 months of their release, including one-third who used
emergency rooms and one-fifth who were hospitalized.
− Although returning prisoners received these health services, their rates of treatment
for specific health conditions deteriorated, suggesting that they received episodic care
for acute problems but that continuous treatment of specific health conditions
suffered. Barriers to continuous care other than cost need to be explored further.

Reentry outcomes varied by health status. Individuals with physical health, mental health,
and substance abuse conditions followed distinct reentry trajectories, reporting
significantly different experiences with regard to housing, employment, family support,
substance use, and criminal involvement. Respondents with physical health conditions, as a
group, had the fewest distinct challenges. At the other end of the spectrum, respondents
with mental health conditions reported poorer outcomes in most domains. In between were
respondents with substance abuse problems, who were more likely to relapse into
substance use and recidivate, but otherwise had differential experiences by gender.

Individuals with physical health, mental health and substance abuse problems need housing
assistance, as they were less likely than other returning prisoners to have made living
arrangements before release. Mentally ill men and women and substance abusing women
had particular difficulties with homelessness after release.

Individuals with physical health, mental health and substance abuse problems reported
poorer employment outcomes compared to other returning prisoners. Our findings suggest
two different pathways:
− Respondents with physical and mental health conditions were in poorer health and felt
that health problems interfered with their ability to work.
− Respondents with substance abuse problems did not have greater physical limitations,
but reported more criminal behavior, having been more likely to earn money through
illegal activities.

Family members provided much tangible and emotional support, being the primary source
of postrelease housing, as well as an important source of financial support. Mentally ill
men and women received relatively lower levels of family support compared to other
returning prisoners, as did substance-abusing women. Men with substance abuse problems,
on the other hand, received higher levels of housing and financial support from family

Health and Prisoner Reentry

members compared to other returning men. (Physical health status did not influence family
support.)

In some instances, the relationship between health status and reentry outcomes differed by
gender. This was particularly evident in the provision of family support to respondents
with substance abuse problems: substance abusing men received more tangible assistance
than other men, yet substance abusing women received less help than other women. Since
women typically experienced poorer outcomes than men regardless of health status, the
interaction between gender and health status amounted to a double disadvantage and a
large magnitude of difference. The shares of substance abusing men and women living
with family, for example, differed on the order of 20 percentage points.

Eight to ten months after release, about one-third of returning prisoners reported recent
substance use. Men and women with preprison substance abuse problems were more likely
to use again after release. Interestingly, men with physical health conditions were less
likely to engage in substance use after release. (Returning prisoners with mental health
conditions and women with physical health conditions did not report differential postprison
substance use.)

One in five returning prisoners were back in state prisons within a year of their release.
Substance abusers were more likely to have engaged in criminal behavior and more likely
to have been reincarcerated. We did not observe consistent relationships between physical
and mental health, criminal behavior, and reincarceration rates. Nonetheless, having any
type of health condition—physical, mental, or substance abuse—was associated with either
engaging in more criminal activity or having a higher likelihood of reincarceration.

POLICY IMPLICATIONS

This research has demonstrated how the transition from prison to community differs for
individuals with health problems, and knowledge of these specific differences can be used to
target interventions to improve reentry outcomes. However, practitioners and policymakers must
first recognize the scope of the problem. This study took a cross-disciplinary look at a
representative sample of returning prisoners and found that (a) nearly all had some type of
physical health, mental health, or substance abuse problem and (b) a sizeable fraction had
multiple types of health conditions. Policymakers and practitioners would be well served to
adopt a new paradigm that recognizes health as a universal rather than a special needs concern
among returning prisoners.
Given that health problems influence reentry outcomes, and that nearly all returning
prisoners have health issues, an assessment of health needs should be part of each individual’s
reentry planning process. Prison medical records are not sufficient to gauge health needs, as
many did not utilize prison health services while incarcerated. A prerelease health assessment
could be as simple as a self-administered checklist to screen for problems potentially requiring
follow-up care or management.

Health and Prisoner Reentry

Returning prisoners with physical health conditions, as a group, had the fewest distinct
challenges. Their greatest needs after release were employment and income support, as
health problems often impeded their ability to work. Two strategies are indicated.
− Employment outcomes might be improved if health issues were addressed early in the
postrelease period so that health conditions would not deteriorate.
− Nonetheless, a share of returning prisoners with health problems will require nonwork
sources of financial support. Applications for disability and other forms of public
assistance should be submitted during prison to minimize the amount of time between
release and receipt of services.

Returning prisoners with mental health conditions experienced reentry difficulties across a
range of domains: they had poorer housing and employment outcomes and reported higher
levels of postrelease criminal involvement. This multitude of difficulties, combined with
lower levels of family support, suggests that an integrated case management approach may
be beneficial.

Returning prisoners with substance abuse problems, regardless of gender, engaged in more
postrelease substance use and criminal behavior than other returning prisoners and were
more likely to have been reincarcerated within one year of release. Increasing the amount
of treatment available to substance abusing men and women may serve to reduce aggregate
rates of postrelease use.

Health and Prisoner Reentry

Family members provided much help to returning prisoners, and though the level of support
varied, the majority of individuals with health conditions received some assistance from family
members. Discharge planning should consider the role of family members, including any risks
posed by specific individuals, and realistically assess the amount of support they can provide by
including them in the planning process whenever possible.
Nonetheless, policymakers and practitioners should remember that a sizeable minority—
one-fifth of men and one-quarter of women—did not receive any tangible assistance from family
members. Certain subgroups are more likely to need services from governmental or charitable
sources. Regardless of specific health conditions, women generally experienced poorer outcomes
than men, often stemming from less supportive relationships with family members. Men and
women with mental health conditions and women with substance abuse problems also received
lower levels of support relative to other men and women.
In conclusion, returning prisoners with physical health, mental health, and substance abuse
conditions had reentry experiences that differed in many ways from the “average” returning
prisoner. Our typologies of reentry experiences and service needs are intended to inform
policymakers and practitioners about the types of issues that returning prisoners with health
problems confront. While informative, this cannot take the place of individual, client-centered
discharge planning, especially since so many returning prisoners reported multiple types of
health conditions. The actual services provided to individual returning prisoners must be targeted
to their particular needs.

Health and Prisoner Reentry

1 Introduction

Each year, nearly 700,000 men and women are released from prison into communities across the
United States (Sabol, Minton, and Harrison 2007). Many do not make a successful transition:
two-thirds are arrested within three years and one-half are returned to prison, either for parole
violations or new crimes (Langan and Levin 2002). This revolving door phenomenon is costly in
terms of criminal justice system expenditures, not to mention the human and social costs borne
by crime victims, returning prisoners, their families, and the communities in which they reside.
Since 2001, the Urban Institute has been conducting research to understand the process of
prisoner reentry and identify factors associated with reentry success and failure, with the goal of
informing policy and practice. This study, Returning Home: Understanding the Challenges of
Prisoner Reentry, provides a rich description of the reentry process through a longitudinal series
of interviews with a representative sample of returning prisoners before and after their release
(see Methodology section for details). In this report we describe the health status of returning
men and women and examine the role that health problems and health treatment play in shaping
reentry outcomes. This study takes a wide view of “health” and includes physical health
problems, mental illness, and substance abuse in the analysis.
The results presented here corroborate other research documenting high rates of physical
health conditions, mental illness, and substance abuse in correctional populations (National
Commission on Correctional Health Care [NCCHC] 2002). Returning Home respondents
typically had one or more chronic health conditions at the time of their release. The majority of
men and women in the study sample had chronic physical and mental health conditions at the
time of their release from prison. About 7 in 10 also reported levels of preprison substance use
consistent with abuse and dependence. Including substance abuse under the rubric of health, we
found almost none without health concerns: more than 4 out of 5 men and women had at least
one physical health, mental health, or substance abuse problem at the time of their release.
Available treatment, however, did not match the level of need in prison, and treatment rates
declined once individuals were released to the community.
This report describes the ways in which returning prisoners with physical health conditions,
mental health conditions, and substance abuse problems navigated the first year after release and
demonstrates the distinct challenges they faced with regard to finding housing and employment,
reconnecting with family members, abstaining from substance use and crime, and avoiding a
return to prison. Anecdotal evidence suggests several reasons why prisoners with health
problems often have a more difficult reentry process than others (Travis 2005). Returning
prisoners face multiple, often simultaneous tasks as they embark on the process of reestablishing
their lives outside prison—finding housing, getting a job, having enough money to live on,
reconnecting with children and family—and these intermediary steps influence the ability to live
a drug-free and crime-free life. Unresolved health and substance use problems often complicate

Health and Prisoner Reentry

an already challenging transition. Returning prisoners with health problems may be unable to
engage in work or other activities because of pain or sickness, and their families may be
unwilling or unable to serve as a fallback support. They are additionally confronted with the
tasks of managing their health problems, such as accessing health care and keeping up with
medications or appointments. Those with severe or unmanaged health problems face an
increased risk of adverse outcomes, including physical illness, relapse into drug use or,
particularly in the case of mental illness, inappropriate behavior that provokes a police response.
It stands to reason that successful treatment of returning prisoners’ health conditions could
increase their chances of reentry success by improving their ability to work, support themselves,
and abstain from substance use, all of which have been shown to contribute to desistance from
criminal activity.
Addressing the health problems of returning prisoners has the potential to improve
individual health and reentry outcomes. The benefits may also extend beyond the individual to
the communities in which returning prisoners reside. Released prisoners return in relatively high
concentrations to a small number of socioeconomically disadvantaged communities in America’s
urban centers (Lynch and Sabol 2001). For example, Houston received one-quarter of all persons
returning from Texas prisons, and 25 percent of those returned to just seven neighborhoods
(Watson, Solomon, La Vigne and Travis 2004). Given the extent to which many individuals
cycle in and out of correctional facilities, former prisoners comprise a respectable share of the
population in certain communities. This concentration in some of the most disadvantaged urban
areas has created a public health opportunity whereby attending to the health needs of prisoners
and former prisoners may affect the course of a number of epidemics. Research has shown that
sizeable portions of the total number of Americans with HIV, tuberculosis, and hepatitis, for
example, serve time in correctional facilities each year (NCCHC 2002). If individuals are
engaged in treatment, either in prison or after release, there is the potential to reduce the burden
of illness and prevent further disease transmission.
In the sections that follow, we present a description of returning prisoners in the study
sample and an overview of their reentry experiences regardless of health status. Following the
overview, we address the experiences of those with physical health conditions, mental health
conditions, and substance abuse problems individually: we describe the prevalence of each type
of condition and the extent to which individuals received treatment, and then proceed to discuss
the ways in which those with physical, mental, and substance abuse problems had distinctly
different reentry experiences from other returning prisoners. The report concludes with a
discussion of the role of health in reentry and offers empirically-based recommendations for
improved policy and practice.

Health and Prisoner Reentry

2 Prisoner Reentry: An Overview

PROFILE OF RETURNING PRISONERS

Returning Home study participants were representative of released prisoners in their states who
had been sentenced to at least one year in prison and planned to return to their states’ major
metropolitan area (see Methodology section for further detail). As shown in Table 1, respondents
were 36 years old, on average, at the time of release. The majority were from minority racial
groups, with most men and about two-thirds of women describing themselves as being black or
of another (nonwhite) race. About 1 in 10 respondents reported Hispanic ethnicity and almost all
were U.S. citizens.

Men and women in the study typically served this prison term for a new crime as opposed to
a parole violation (about one-third), with drug offenses being the most common conviction
charge. Beyond these general characteristics, however, the men and women in our sample had
different criminal justice profiles, consistent with the characteristics of the Ohio and Texas
release populations. About one-third of men were serving time for a drug offense, with the
remainder divided evenly among violent, property and other offense types. Over half the women
were serving time for a drug offense and the remainder was divided evenly among property and
other offense types; unlike men, very few women (2 percent) were serving time for violent
offenses. Due to the relatively lower seriousness of their offenses, women served shorter
sentences, an average (median) of 8 months compared to the 18 months men served. Since
shorter sentences are less likely to include a term of parole supervision, only one-third of women
expected to be under parole supervision upon release compared to two-thirds of men.
HEALTH STATUS OF RETURNING PRISONERS

Soon-to-be released prisoners typically had one or more health conditions. The majority of men
(54 percent) and women (77 percent) in the study sample had chronic physical and mental health
conditions at the time of their release from prison. (Many chronic health conditions in
correctional populations are typically not acquired during incarceration, but are present before
prison admission [NCCHC 2002; Centers for Disease Control and Prevention 2006a]). About 7
in 10 also reported levels of preprison substance use consistent with abuse and dependence.
Including substance abuse under the rubric of health, we found almost none without health
concerns, and many with multiple, overlapping types of health conditions. Although substance
abuse is not often included in discussions of “health,” the high level of substance use in
correctional populations is indeed a health concern. Substance use at the level of addiction is a
mental health condition in its own right, an Axis I disorder according to the Diagnostic and
Statistical Manual of Mental Disorders (American Psychological Association 2000), and may be
viewed as a chronic medical illness (McLellan, Lewis, O’Brien, and Kleber 2000). Moreover,
Health and Prisoner Reentry

regardless of addiction, the physiological effects of drugs and alcohol are damaging to overall
physical health. In addition to the risk of overdose, substance use increases the likelihood of
developing other chronic conditions, including cardiovascular disease and cirrhosis of the liver
(National Institute on Drug Abuse [NIDA] 2004), and contributes to the transmission of several
infectious diseases, most notably HIV.
Table 1. Selected characteristics of Returning Home respondents
Men

Women

(n=838)

(n=262)

Demographics
Age (median)

36 years

Black or African American race**

67.2%

53.3%

White or Caucasian race**

19.5%

34.0%

Hispanic ethnicity**

12.2%

11.3%

High school diploma or GED at release*

69.7%

59.8%

Employed at any time during the 6 months before prison**

72.5%

54.5%

Married or living together as married before prison

24.0%

23.8%

Had children under age 18 at release**

48.3%

65.4%

Lived with minor children before prison**

29.6%

38.8%

Number of close family relationships before prison (median)

Number of close family relationships during prison (median)

Education and employment

Family relationships

Criminal history
Age at first arrest (median)**

17 years

21 years

Served time in a juvenile facility**

36.0%

20.9%

Previously convicted of a crime

82.3%

80.2%

Served time in prison before**

66.0%

56.9%

Had parole or probation revoked before**

48.9%

49.6%

18 months

8 months

Incarcerated in state jail **

24.1%

72.5%

Violent offense conviction**

22.8%

2.3%

Property offense conviction

20.5%

21.1%

Drug dealing conviction

11.1%

16.9%

Drug possession conviction

24.7%

38.7%

Current term resulted from probation or parole violation

35.6%

34.5%

Will be released to parole supervision**

62.2%

31.4%

Current prison sentence
Time served (median)**
a

aSome Texas respondents served their sentences in state jails. These are unique to Texas and distinct from local or
county jails in that they are part of the state correctional system. Unlike state prisons, however, state jails exclusively
house low-level felony convicts with sentences of two years or less. The relatively shorter length of stay in state jails
may have reduced opportunities for service provision during incarceration.

Health and Prisoner Reentry

Despite the high prevalence of health conditions, most men (81 percent) and women (75
percent) described their health as excellent or good while in prison. This dissonance between the
prevalence of health conditions and positive self-assessments is likely due to the nature of many
chronic ailments. Respondents often reported serious yet “silent” conditions, such as diabetes,
that require care but can remain asymptomatic for extended periods of time. Nonetheless, about
one-fifth of men (19 percent) and one-quarter of women (25 percent) assessed their health as fair
or poor, with women being more likely to give their health a negative assessment.
One-half of men (49 percent) and two-thirds of women (67 percent) had chronic physical
health conditions requiring long-term management and care at the time of their release. We
assessed health status during the prerelease interview using respondent self-reports. Respondents
were given a list of chronic physical and mental health conditions and asked which conditions a
doctor or nurse had diagnosed them with. The physical conditions most often reported included
asthma, high blood pressure, and diabetes, but one-fifth of returning prisoners reported having
communicable diseases like hepatitis B and C, tuberculosis, and HIV infection. Among those
with chronic physical health conditions, the majority of men (64 percent) and women (73
percent) reported receiving treatment during prison, but a sizeable share (one-third and onequarter, respectively) did not.
Fifteen percent of men and over one-third of women (35 percent) reported mental health
conditions. As noted above, mental health status was also assessed through respondent selfreports at the prerelease interview and respondents indicated whether they had been diagnosed
with depression or other mental illness. While the actual prevalence is probably double what was
reported, as other researchers have found in this population (James and Glaze 2006), the study’s
definition of mental illness is nonetheless based on respondent self-reports while in prison. (See
Section 4 for the rationale for this decision and for more information on mental health measures
selected for this report.) Of those reporting mental health conditions, 60 percent of men and 57
percent of women reported receiving treatment during prison, meaning about 4 in 10 did not.
Most respondents reported using drugs and alcohol regularly during the six months before
this incarceration, and many had substance abuse problems that remained unresolved at the time
of release. We assessed probable abuse by asking respondents how often they used drugs or
drank to intoxication in the six months preceding their current prison term. The study defined
substance abuse as alcohol intoxication or drug use more often than once a week. This level of
use was correlated with a greater number of social and interpersonal problems related to
substance use (e.g., problems at work, arguments at home) and more signs of addiction (e.g.,
physical tolerance, withdrawal symptoms). By this definition, 7 out of 10 men and women had a
substance abuse problem. Interestingly, there were very few casual users: about 2 in 10
respondents used drugs or drank to intoxication weekly or less often and about 1 in 10 reported
no use at all. (Throughout this report, we use the term substance abuse to refer to preprison use
more than once a week. This is distinct from substance use, which refers to any drug use or
alcohol intoxication.) During prison, 52 percent of men and 41 percent of women with substance
abuse problems received some treatment services—this term is used to encompass participation
in formal substance abuse treatment regimens as well as self-help groups like Alcoholics
Health and Prisoner Reentry

Anonymous (AA) and Narcotics Anonymous (NA)—and 29 percent and 14 percent,
respectively, participated in formal substance abuse treatment programs. Given the chronic
nature of substance abuse, even with treatment, many returning prisoners are at risk of relapse
and require aftercare upon release.
Most respondents (84 percent of men and 92 percent of women) reported at least one
physical health, mental health, or substance abuse problem, and large shares (39 percent of men
and 62 percent of women) had multiple types of health conditions. This is especially relevant to
practitioners, as it is important to recognize that clients or patients presenting with one type of
health problem often have other coexisting conditions. Clinical experience with co-occurring
mental health and substance abuse problems has shown that incomplete attention to one type of
problem decreases the likelihood of successfully treating the others (Quello, Brady, and Sonne
2005). Figures 1 and 2 display two different views of the relative share of each of these
conditions and the overlap among them. Notably, women reported more co-occurring conditions
than men. Among men with physical health conditions, one-fifth also reported mental health
conditions; among women this figure was over one-third. The majority of respondents with
physical health conditions, regardless of gender, also reported substance abuse problems. Among
respondents with mental health conditions, male and female, nearly all had co-occurring physical
and substance abuse problems. Of particular interest are those who are dually-diagnosed with
mental health and substance abuse disorders: we estimate that one-tenth of all returning men and
one-quarter of all returning women fit this description. Among respondents with substance abuse
problems, one-half of men and two-thirds of women also had physical health conditions;
moreover, 16 percent of men and 37 percent of women with substance abuse problems reported
mental health conditions.
Figure 1. Interplay of physical health, mental health, and substance abuse conditions among returning
prisoners, by gender
Most returning prisoners (84% of men and 92% of women) reported having a physical health, mental health, or
substance abuse problem. Many (39% of men and 62% of women) reported more than one type of condition.
Men

Women
70% reported
substance
use more
than once a
week

66% reported
substance
use more
than once a
week
49%
reported a
physical
health
condition

30%

25%

14%

6%
3%

4%
1%

15%
reported a mental
health condition

30%
67%
reported a
physical
health
condition

12%

15%
18%
7%

8%
2%

35%
reported a
mental health
condition

Note: Percentages may not sum exactly because of rounding.
Source: Returning Home original data collection in Ohio and Texas (838 men and 262 women).

Health and Prisoner Reentry

Figure 2. Prevalence of physical health, mental health, and substance abuse conditions among returning
prisoners, by gender
More often than not, returning prisoners with any given health condition reported a second, co-occurring condition.
Men

100%

Women

100%

90%

80%

80%
70%

70%
physical health
condition

60%
50%

mental health
condition

40%
30%

substance
abuse

20%
10%

physical health
condition

60%
50%

mental health
condition

40%
30%

substance
abuse

20%
10%

all three

0%
Physical health
condition

Mental health
condition

Substance
abuse

Physical health
condition

Mental health
condition

Substance
abuse

Source: Returning Home original data collection in Ohio and Texas (838 men and 262 women).

Returning prisoners’ capacity to access community-based care for their chronic health
conditions was limited by a lack of health insurance. As Figure 3 shows, most respondents were
without health insurance in the first two to three months after release. Though respondents may
have had insurance before prison, benefits were likely suspended or terminated during their
incarceration. Medicaid benefits, for example, are suspended during incarceration and the
reinstatement of benefits can take several months. Rates of insurance coverage increased by the
time of the eight to ten month interview, with men reporting private insurance most often and
women reporting Medicaid coverage, but the majority remained uninsured. This results in a
situation where health could deteriorate and hinder reentry success.
Figure 3. Health insurance coverage among returning prisoners, by gender
Most respondents were uninsured after prison release.
Men

100%

80%

Women

100%

78%

80%
68%

66%
60%

60%

58%

2-3 months postrelease
8-10 months postrelease

2-3 months postrelease
8-10 months postrelease
40%

40%

21%
20%

20%

16%

12%
5%

8%
2% 1%

3% 3%

11%
6% 6%
1% 1%

1% 1%

Medicare

VA health
benefits

14%

4% 5%

0%
None

Medicaid

Medicare

VA health
benefits

Private
insurance

Other

None

Medicaid

Private
insurance

Other

Note: Many respondents who reported other insurance said they had a "gold card." This is a local financial assistance
program through the Harris County (Texas) Hospital District that enables individuals to receive free or subsidized
health care.
Source: Returning Home original data collection in Ohio and Texas. Two to three month postrelease measures are
based on 652 men and 202 women. Eight to ten month postrelease measures are based on 539 men and 153
women.

Health and Prisoner Reentry

THE “AVERAGE” REENTRY EXPERIENCE

Returning prisoners face a challenging transition regardless of their health conditions. They must
find housing and a means to support themselves, both of which often depend on their ability to
reconnect with family members and social networks. Maintaining sobriety and refraining from
criminal activity are also important to avoiding a return to prison. In this section, we provide an
overview of respondents’ reentry experiences in the first year related to housing, employment,
family, substance use, and criminal involvement, regardless of health status. (Table 2
summarizes these key reentry outcomes.) The remainder of the report will then elaborate on the
experiences of those with physical health, mental health, and substance abuse problems and
demonstrate, empirically, how persons with each of these health conditions followed reentry
trajectories that differed from the “average” experience.
Table 2. Key reentry outcomes eight to ten months after prison release, by gender
Men

Women

…in the first 2-3 months**

62.6%

50.0%

…in the first 8-10 months**

69.7%

56.4%

Housing
Ever lived with family members

Ever experienced any homelessness
…in the first 2-3 months

5.6%

7.4%

…in the first 8-10 months**

5.9%

13.5%

76.0%

62.3%

…2-3 months after release**

37.3%

30.7%

…8-10 months after release**

53.5%

33.3%

81.6%

73.7%

Current drug use or alcohol intoxication

34.7%

35.9%

Current participation in substance abuse treatment services

20.9%

21.3%

Self-reported crime, arrest, or parole violation

37.9%

46.8%

Reincarcerated within 12 months of release

17.4%

21.0%

Employment
Any employment since release
Currently employed**

Family support
Received any financial support from family members*
Substance use

Criminal involvement

Health and Prisoner Reentry

Housing

Finding a place to live is perhaps the first challenge that returning prisoners face upon release.
When we surveyed respondents in the month before release, nearly one-third had not made
arrangements for a place to live. Nevertheless, the majority expected to live with family
members and this expectation was borne out over the course of the first eight to ten months after
prison. Beginning with the first night out of prison, family was the primary source of housing for
returning men and women, though more so for men than for women.
Over the course of the first eight to ten months after release, returning prisoners reported a
variety of living arrangements, with 40 percent of men and 59 percent of women reporting at
least one change in residence. At the time of the second postrelease interview, the majority of
men (70 percent) and women (56 percent) had spent at least some time living with family
members, with men being more likely than women to have done so. A smaller share of
respondents reported having lived in their own homes or apartments, about 3 in 10 men and 4 in
10 women. Returning prisoners were more reliant on family members for housing during the first
eight to ten months after release than they had been during the six months before this
incarceration. Before prison, the greatest share had lived in their own homes; living with family
was secondary, reported by roughly one-third of men and one-quarter of women.
Many respondents (about 2 in 10 men and 3 in 10 women) reported having trouble keeping
housing since their release. However, respondents reported lower rates of homelessness after
release compared to their experience before this incarceration.1 During the six months before this
prison term, 14 percent of men and 27 percent of women had been homeless, defined as having
no set place to live, living on the streets, or staying in shelters or rooming houses. Reports of
homelessness in the first year after prison were much lower: 5 percent of men and 7 percent of
women reported being homeless on the first night out, and rates of homelessness remained at this
level throughout the first two to three months postrelease. Eight to ten months after prison,
homelessness among men remained at a similar level but 13 percent of women now reported
having been homeless for at least some time since release. Nonetheless, this is considerably
lower than the share that reported homelessness before prison.
While most respondents’ housing needs were met, at least at a basic level, there is a caveat.
Men and women often lived with individuals who could jeopardize their reentry success, be they
family members, intimate partners, friends, or acquaintances. Two to three months after release,
over one-quarter of men and over one-third of women reported living with former prisoners and
current substance users. Eight to ten months after prison, the share of men living with such
individuals remained about the same. Among women, however, the share living with former
prisoners and current substance users increased to 4 in 10. In addition to the risk of antisocial

Postrelease homelessness may be underestimated because it was measured using a calendar that asked respondents
to choose one category that best described their housing situation in each postrelease month. Shorter episodes of
homelessness may not be adequately captured by this method.
Health and Prisoner Reentry

influences, such a housing situation constitutes a violation of release conditions that could result
in a return to prison for those on parole supervision.
Employment

Finding and maintaining employment are critical dimensions of reentry. Research has shown that
employment and higher wages are associated with lower rates of criminal activity (Bernstein and
Houston 2000; Western and Pettit 2000). At the same time, returning prisoners face many
obstacles in gaining legitimate employment, including low levels of education, limited
vocational skills and work experience, and a reluctance on the part of employers to hire exprisoners (Harlow 2003; Holzer, Raphael, and Stoll 2004).
About one-half of men had found some employment within the first 2 to 3 months after
release; this figure rose to three-quarters by the time of the 8 to 10 month postrelease interview.
Employment rates were lower among women but followed a similar pattern: roughly 4 in 10
found some employment in the first 2 to 3 months and this figure rose to 6 in 10 by the time of
the 8 to 10 month postrelease interview. However, not all those who found jobs were able to
maintain employment. Eight to ten months after release, just over half the men and one-third of
women reported they were currently employed. Men generally reported better employment
outcomes than women, and had worked for more postrelease months.
Returning prisoners supported themselves through a combination of means that included
family and friends, legal employment, casual or “under the table” work, public assistance, and
illegal activities. Family and friends, legal employment, and “under the table” work were the
most commonly reported sources of financial support. Comparing respondents’ pre- and
postprison circumstances, we found that men and women were less self-sufficient 8 to 10 months
after release than they had been before this incarceration and reliance on personal social
networks increased. Family and friends were particularly important early in the postrelease
period and served as the most common source of income two to three months after release, cited
by nearly two-thirds of men and women. Eight to ten months after release, more returning
prisoners supported themselves through work, but family and friends continued to be an
important source of support. One-half of men supported themselves through legal employment,
but family and friends were the second most common source of income, reported by 4 in 10; by
contrast, two-thirds of men had supported themselves through a job before prison. Among
women, family and friends remained the predominant source of income, reported by 6 in 10,
with relatively fewer women (3 in 10) supporting themselves through legal employment; before
prison, one-half of women had supported themselves through a job.
Government programs and illegal activities were important sources of income for smaller
shares. Public assistance programs, such as food stamps and housing aid, provided support to
one-fifth of men and women. Disability programs like Social Security Disability Insurance
(SSDI) and Supplemental Security Income (SSI) also provided income, but more so to women
(18 percent) than men (6 percent). Finally, illegal activities were sometimes a source of income,

Health and Prisoner Reentry

reported more often by women (15 percent) than men (6 percent). All told, returning men
reported an average (median) monthly income of $950; among women this figure was $700.2
Family support

Respondents frequently described their family relationships as close and supportive and, as noted
in the previous sections, family members often provided practical support to returning prisoners,
including housing and financial support. Research has shown that strengthening the family
network and maintaining contact can improve outcomes for returning prisoners (Sullivan, Mino,
Nelson, and Pope 2002).
The Returning Home study used a broad definition of family, consistent with ethnographic
accounts of inner-city neighborhoods (LeBlanc 2003). Family members included blood or legal
relatives, guardians, people with whom the respondent had a child in common, and intimate
partners. On average, men and women reported having 4 to 5 close family members before
prison, and this number of close relationships remained consistent during their incarceration and
after release. Men and women typically cited their mothers and sisters, respectively, as their
closest family members upon release. Respondents reported high levels of emotional support and
closeness before prison: on a scale of one to four, four being the highest, men and women rated
family support above three, though women tended give lower ratings than men. Respondents
perceived similar levels of emotional support and closeness during their incarceration, and
ratings of support increased after release.
The majority of returning prisoners expected and received tangible support from their
families upon release. Eight to ten months after release, 82 percent of men and 74 percent of
women reported having received some tangible help from family members. As noted earlier,
family was the number one source of housing for returning prisoners in the first year after
incarceration, with 70 percent of men and 56 percent of women having lived with a family
member for some time in the first 8 to 10 months after release. Family members also provided
other tangible assistance, including food, clothing, money, personal items, and transportation.
Returning prisoners, on average, found family members to be more helpful than they had
expected.
While family members provided much needed assistance, family influences are not
universally positive. Returning prisoners often reported familial problems with delinquency and
substance abuse. The majority had relatives with criminal convictions and addiction problems,
and sizeable shares (3 in 10 men and 4 in 10 women) had relatives who were currently in prison.
Family violence is also a concern: 1 in 10 men and 3 in 10 women reported having been
threatened or victimized by a family member in the year before this incarceration. Conversely,
13 percent of men and 7 percent of women reported they had threatened or victimized family
members.

Note, however, that the difference in means was not statistically significant.
Health and Prisoner Reentry

Finally, it is important to remember that while the majority of returning prisoners expected
and received help from their families, there was a sizeable minority that did not. Eight to ten
months after prison, one-fifth of men and one-quarter of women had not received any tangible
assistance from family members, with women being less likely than men to have lived with
family members or received financial support.
Postrelease substance use

Substance use puts returning prisoners at risk for a number of adverse outcomes. The possession
of illegal drugs alone puts individuals at risk for arrest, parole violation, and reincarceration.
Moreover, substance use may propel individuals to commit other crimes in order to finance drug
purchases. Substance use also affects other outcomes in that it can impede family relationships,
employment, and health. When discussing substance use, we also include alcohol intoxication,
since abstinence from alcohol is frequently a condition of parole release. Furthermore, alcohol
abuse has many of the same adverse consequences as drug use in that it increases the risk of
violent behavior, disrupts family relationships and employment, and has ill effects on health.
Many respondents reported postrelease substance use and this was more common among
women than men. Eight to ten months after prison, one-third of men and nearly half the women
reported drug use or alcohol intoxication since their release. The most common substances used
postrelease by men were alcohol (to intoxication), marijuana, and cocaine, respectively. Cocaine
was the most common substance used postrelease by women, reported by about one-quarter,
followed by marijuana use and alcohol intoxication. Current use, measured as drug use or
alcohol intoxication during the month of the interview, increased among men during the
postrelease period from about one-quarter 2 to 3 months after prison to one-third 8 to 10 months
after release. Among women, however, current use remained constant throughout the postrelease
period, with about one-third reporting use at each postrelease interview. Most of the respondents
who reported substance use after prison reported illegal drug use, with women being more likely
than men to have used illegal drugs and cocaine in particular. Women’s postrelease use was also
more likely than men’s to be characterized by abuse and dependence. Postprison substance use
was reflective of preprison substance abuse patterns, except that reported rates of use were
lower; it is likely that respondents underreported the actual level of use since we were asking
about current rather than past behavior, and many were under parole supervision.
Substance abuse treatment services did not match the level of need during prison and
treatment rates fell further after prison release. During prison about half of men and 4 in 10
women participated in substance abuse treatment programs or self-help groups like AA or NA.
Two to three months after prison, just one-quarter of men and 3 in 10 women were participating
in such services. Eight to ten months after prison only one-fifth were participating in such
services. Most of the respondents in recovery after prison were in self-help groups like AA and
NA; fewer than 10 percent reported participating in formal drug or alcohol treatment programs.
Criminal involvement

One-fifth of returning men and women were back in state prisons within a year of their release,
according to official records from the Ohio and Texas Departments of Corrections. More often
Health and Prisoner Reentry

than not, they were reincarcerated for new crimes rather than for parole violations. While there
was no gender difference in the overall reincarceration rate, women were more likely than men
to be back in prison for new crimes, probably because women were less likely than men to have
been under parole supervision upon release.
While overall reincarceration rates did not differ by gender, returning men and women
reported different levels of engagement in postrelease criminal activity. Eight to ten months after
prison, nearly one-fifth of men and one-third of women reported having committed crimes since
release. Similar shares reported having been arrested, with roughly one-tenth having spent at
least one postrelease month in jail. Among those on parole, roughly one-third reported violating
conditions of their release. All told, 38 percent of returning men and 47 percent of women
recidivated, as measured by a combination of self-reported crimes, arrests, and parole violations.
(This was not a statistically significant gender difference.) The most common crimes reported
were drug possession, drug selling, robbery and theft. Women also reported “other” crimes,
which would likely include prostitution, and were more likely than men to report that illegal
activities were a source of postrelease income.

Health and Prisoner Reentry

3 Physical Health and Reentry

PHYSICAL HEALTH STATUS AND TREATMENT

At the time of prison release, one-half of men and two-thirds of women reported having been
diagnosed with a chronic physical health condition. Physical health status was assessed by asking
respondents to note those conditions that a doctor or nurse had diagnosed them with. Table 3
shows the full range of conditions that respondents were asked about and their self-reported rates
of illness; for comparison purposes, these are displayed alongside national prevalence estimates
for correctional populations developed by the National Commission on Correctional Health Care.
Asthma, hepatitis infection, and high blood pressure were the top three conditions reported. Men
most commonly reported having high blood pressure (20 percent), hepatitis (11 percent), asthma
(10 percent), high cholesterol (8 percent), and arthritis (7 percent). Women reported similar
ailments, with asthma (25 percent), high blood pressure (23 percent), hepatitis (15 percent), back
pain (15 percent), and arthritis (14 percent) being the most prevalent conditions. Despite the high
prevalence of health conditions, most had positive feelings about their health, describing it as
excellent or good while in prison. This is plausible, since many of these chronic conditions can
remain asymptomatic for long periods of time, particularly if treated appropriately. Nevertheless,
one-quarter of men and one-third of women with physical health conditions assessed their health
as fair or poor in prison.
Chronic communicable diseases are of particular importance to the public health since,
without intervention, they can be transmitted to other prisoners, correctional staff, and the
families and communities to which prisoners return. One-fifth of returning prisoners reported
having been diagnosed with a chronic communicable disease (i.e., hepatitis, tuberculosis, and
HIV). Hepatitis B or C infection (11 percent) was the most common communicable disease
reported by men, followed by tuberculosis infection (5 percent), and HIV or AIDS (2 percent).
Women reported similar rates of hepatitis (15 percent) and tuberculosis (4 percent), but were
more likely to report having HIV or AIDS (6 percent).
For the most part, the rates of illness reported by respondents were comparable to national
estimates. Reported rates of some serious communicable diseases, notably hepatitis and
tuberculosis infections, were lower than the national estimates. These differences may reflect
regional variation in the prevalence of these diseases. It may also be that respondent self-reports
underestimate the true prevalence of disease because many serious conditions can be
asymptomatic and remain undetected without proper screening and medical care.

Health and Prisoner Reentry

Table 3. Physical health conditions among returning prisoners

Any physical health condition**

Male Returning Home
respondents (n=831)

Female Returning Home
respondents (n=252)

U.S. correctional
population

48.5%

66.7%

—

Arthritis**

7.3%

13.9%

—

Asthma**

10.3%

25.0%

8.5%

Back pain**

6.3%

15.1%

—

Cancer

2.0%

3.2%

—

Chronic lung disease**

2.8%

7.5%

—

Diabetes

5.4%

6.0%

4.8%

Heart disease
Hepatitis B or C

5.5%

5.6%

—

11.2%

15.1%

Hepatitis B, 2.0%
Hepatitis C, 17.0–18.6%

High blood pressure

20.1%

23.4%

18.3%

High cholesterol

8.1%

5.6%

—

HIV or AIDS

2.0%

6.3%

HIV, 2.3-2.98%

Stroke

1.3%

1.2%

Tuberculosis

4.7%

4.0%

AIDS, 0.5%
—
Latent infection, 7.4%
Active disease, 0.04%
Note: T-tests were conducted to identify statistically significant gender differences. A single asterisk (*) denotes
differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with a p-value ≤ 0.05. Significance
testing for gender differences was restricted to Texas participants (409 men and 252 women) to control for state
effects; these statistics are available from the authors upon request.
Sources: Returning Home original data collection in Ohio and Texas. Prevalence estimates of physical health conditions
in the U.S. correctional population are from Chapter 3 of The Health Status of Soon-to-be Released Prisoners, Volume
1 by the National Commission on Correctional Health Care, 2002. Dashes indicate that data on certain health
conditions were not available.

Respondents with physical health conditions often reported co-occurring mental health or
substance abuse problems (Figure 2, shown previously). Among men with physical health
conditions, about one-fifth also reported having a mental health condition and two-thirds
reported preprison substance abuse. Among women, one-third of respondents with physical
health conditions reported also having been diagnosed with a mental health condition and over
two-thirds reported preprison substance abuse. Physical health problems in correctional
populations are often related to substance use. Diseases like HIV and hepatitis are transmitted
through injection drug use and risky sexual behaviors, like bartering sex for drugs. Moreover,
long-term substance use can have damaging physiological effects. Respondents with physical
health conditions were older and reported more years of substance use over their lifetimes,
though, in general, they were no more or less likely to have been active abusers in the six months
before this prison term. Recent cocaine use, however, was correlated with having physical health
conditions, as was heroin use among men.
The majority of returning prisoners with physical health conditions—about two-thirds of
men and three-quarters of women—received some treatment during prison. It is worth noting,
however, that one-third of men and one-quarter of women with identified health conditions did
not report receiving treatment while incarcerated. As shown in figure 4, treatment rates in prison
were variable, depending on the type of health condition. On the high end, treatment rates were
Health and Prisoner Reentry

above 70 percent. Most respondents with diabetes, HIV, and high blood pressure reported being
treated for those conditions. Other conditions, however, had treatment rates below 50 percent,
including back pain, hepatitis, and tuberculosis. The treatment of chronic communicable diseases
is of particular concern because of the potential for disease transmission in the larger community,
yet we observed variable treatment rates. Although treatment rates for HIV exceeded 80 percent,
treatment rates for hepatitis and tuberculosis were considerably lower, between 20 and 40
percent. These findings point to a need for focused research on in-prison health treatment, as it is
difficult to determine whether these treatment rates are medically appropriate within the scope of
this study. Hepatitis and tuberculosis infections both have long periods of latency and treatment
decisions are subjective. For example, tuberculosis infection is treated for 9 months, if at all,
depending on the patient’s age, when he or she became infected, and other clinical factors, such
as evidence of liver damage (Centers for Disease Control and Prevention [CDC] 2006b).
Similarly, viral hepatitis has a lengthy incubation period (20 to 40 years) and clinical guidelines
recommend treatment within a specific window period (CDC 2003).
Figure 4. Percentage of respondents with selected physical health conditions who received treatment, by
gender
Treatment rates varied by the type of health condition and typically dropped upon release from prison.
Men

Women

Asthma

Diabetes

Hepatitis B or C

High blood pressure

HIV or AIDS

HIV or AIDS
during prison
2-3 mo. post
8-10 mo. post

Tuberculosis
0%
Any physical condition

during prison
2-3 mo. post
8-10 mo. post

Tuberculosis

Any physical condition
0%

20%

40%

60%

80%

100%

20%

40%

60%

80%

100%

Source: Returning Home original data collection in Ohio and Texas. Treatment rates are based on the number of
respondents reporting each health condition at each study time point. Sample sizes for each item are available
from the authors upon request.

Even when individuals receive adequate and appropriate treatment services in prison, they
face several barriers to care upon release, including a lack of health insurance and insufficient
linkages to community-based care (Hammett, Roberts and Kennedy 2001). Returning prisoners
are typically uninsured upon release since Medicaid benefits are suspended during incarceration
and the restoration of eligibility can take several months. In the Returning Home study, we found
that many respondents utilized health services despite low rates of insurance coverage. At the
same time, treatment for chronic conditions declined, suggesting that respondents often sought
emergency care for acute health problems.
The majority of respondents with physical health conditions (64 percent of men and 57
percent of women) were uninsured eight to ten months after prison, though they were more likely

Health and Prisoner Reentry

to be insured compared to other returning prisoners. Despite a lack of health insurance, the
majority (over 7 in 10) received some health care during the first year after prison. One-third of
men and women reported emergency room visits and about one-fifth had been hospitalized.
Respondents with physical health conditions were more likely than others to have utilized health
care, particularly in the first 2 to 3 months after release.
Although respondents utilized health services after release, it appears that they sought care
for acute episodes of illness and that continuity of care eroded. Returning Home results show a
decline in treatment rates for specific health conditions once prisoners were released (Figure 4).
Eight to ten months after release, just 48 percent of men and 60 percent of women reported
receiving care for their chronic physical conditions compared to 64 percent and 73 percent,
respectively, who received such care during prison. Looking at prescription drug use in
particular, we found that most prisoners who used prescription medicines during prison received
a supply of medication upon release. However, just two-thirds continued to use that same
medication two to three months after release, and this figure was closer to one-half 8 to 10
months after release. The primary reasons for discontinuing medication, interestingly, were not
cost or access issues. Respondents who stopped using the medications they had taken during
prison most often cited “other” reasons. Thereafter, the most common reasons were that a doctor
in the community determined that the medication was no longer necessary and, alternately, that
respondents themselves decided they no longer needed the medication. This points to a different
set of barriers to continuous care in the community, including patient education, trust, and
coordination between prison- and community-based medical providers.
REENTRY EXPERIENCES OF PRISONERS WITH PHYSICAL HEALTH CONDITIONS

While many returning prisoners had physical health conditions requiring care and management,
the degree to which these affected daily living varied. Despite reporting physical health
conditions, the majority of respondents rated their health as excellent or good during prison.
However, health status worsened over the course of the first postrelease year, particularly for
those with physical health conditions: eight to ten months after release, one-third of men and
one-half of women rated their health as fair or poor (up from one-quarter and one-third,
respectively, during prison). One-third of respondents with physical health conditions reported
that their health problems had limited their daily activities and ability to work; moreover, 14
percent of men and 22 percent of women with physical health conditions had problems so severe
that they qualified for and received disability pensions. In the sections that follow, we examine
the reentry experiences of respondents with physical health conditions and contrast them with
healthier returning prisoners. (See the Methodology section for details on the analytic design.)
Statistically different outcomes with regard to housing, employment, family support, postrelease
substance use and recidivism are detailed below.
Housing and physical health

Returning prisoners with physical health conditions were less likely than others to have had
housing lined up when we surveyed them in the month before prison release. Nonetheless, in
many respects, their postrelease housing experiences were similar to other returning prisoners,
Health and Prisoner Reentry

even on the first night out of prison. Over the course of the first 8 to 10 months out of prison,
those with physical health conditions most often stayed with family members and at rates similar
to other returning prisoners (about 7 in 10 men and 6 in 10 women). They were also similar with
respect to living in their own homes, and their rates of homelessness were no different from
those reported by other returning prisoners.
However, those with physical health conditions were more likely to have trouble keeping
housing and reported moving around more often than other returning prisoners. For men with
physical health conditions, this residential mobility was greater within the first 2 to 3 months of
release, but they were no different from others later, 8 to 10 months after release. The opposite
was observed among women with physical health conditions: 2 to 3 months after release, they
were similar to other returning women, but their residential mobility was greater later on, 8 to 10
months after release, suggesting that men and women may require housing assistance at different
times in the postrelease period. Additionally, women with physical health conditions were more
likely to report living with potentially negative influences (i.e., other former prisoners and
current substance abusers) later in the postrelease period.
Employment and physical health

Many respondents with physical health conditions—about one-third, regardless of gender—felt
their health problems were severe enough to limit their ability to work or partake in other routine
activities. These perceptions were supported by the finding that, eight to ten months after release,
men and women with physical health conditions had less employment success relative to other
respondents (Figure 5). Though the majority (71 percent of men and 58 percent of women) had
worked for at least some time since release, these rates were significantly lower than the
employment rates reported by healthier respondents. Interestingly, men and women with
physical health conditions had found employment at rates similar to other respondents two to
three months after release, suggesting that some months elapse before physical health issues
impede employment outcomes.
While both men and women with physical health conditions had less success in gaining
employment, the effect of physical health on employment and income was more pronounced
among men than it was among women. Returning prisoners supported themselves through
multiple means, as shown in Figure 6. Unlike other men, those with physical health conditions
most commonly reported family and friends as a source of income eight to ten months after
release. They were significantly less likely to support themselves through employment and more
likely to receive financial support through government programs, including Social Security
disability and other forms of public assistance (e.g., food stamps). While women with physical
health conditions had also been less likely to find employment, they were not much different
from other women in terms of how they supported themselves: family and friends were the most
commonly cited source of income, with just one-fifth reporting income through employment.
Women with physical health conditions, however, were more likely to receive support from
disability programs. All told, men with physical health conditions reported a median monthly
income of about $900 whereas women reported about $650. These amounts were somewhat

Health and Prisoner Reentry

lower than the median income reported by others without physical conditions, but the difference
in means was not statistically significant.
Figure 5. Percentage of respondents who were ever employed, by physical health status and gender
Eight to ten months after release, returning prisoners with physical health conditions were less likely than others to
have found a job, despite having found employment at similar rates earlier in the postrelease period.
100%

100%

Men
no physical
illness

80%

Women

80%

no physical
illness

60%

physical
illness

40%

20%

physical
illness

0%
Before prison

2-3 months
postrelease

8-10 months
postrelease**

Before prison**

2-3 months
postrelease

8-10 months
postrelease*

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 808 men and 248
women. Two to three month postrelease measures are based on 657 men and 190 women. Eight to ten month
postrelease measures are based on 536 men and 148 women.

Figure 6. Sources of income eight to ten months after prison release, by physical health status and gender
Physical health had a more pronounced influence on men’s employment and income. Returning men with physical
health conditions were less likely than other men to earn income through legal employment, but the same pattern did
not hold true for women.
Men

Women
43%

Legal employment**

29%
28%

Legal employment
54%

31%
33%

"Under the table" work

45%
42%

Family and friends

61%
60%

Family and friends

10%

SSI/SSDI disability**

24%
18%

"Under the table" work

19%

SSI/SSDI disability**

2%
24%

Public assistance**

physical illness
no physical illness

13%
6%
7%

Illegal activities

20%
18%

Public assistance

17%
12%

Illegal activities

20%

40%

60%

80%

100%

physical illness
no physical illness

20%

40%

60%

80%

100%

Health and Prisoner Reentry

Family support and physical health

Men and women with physical health conditions were no different from other returning prisoners
in terms of receiving emotional and tangible support from their family members after release.
They were as likely as other returning prisoners to have lived with family members in the first 8
to 10 months after release and were similarly likely to have received financial assistance from
their families. However, family relationships posed greater risks to individuals with physical
health conditions, who were more likely to be involved in domestic violence compared to others.
Men with physical health conditions reported having being victimized by family members at
higher rates than other men before prison. Conversely, women with physical health conditions
were more likely than other women to report threatening or hurting family members postrelease.
Substance use and physical health

Men and women with physical health conditions had different experiences with respect to
substance use after prison. Men with physical health conditions tended to have better outcomes
than other men with respect to postrelease use and treatment, whereas women with physical
health conditions were not much different from other women.
Men with physical health conditions reported lower levels of postprison substance use
compared to other men, despite similar levels of use before prison (Figure 7). They reported
fewer months of postrelease substance use and more months of treatment compared to other
men. Two to three months after prison, 2 in 10 reported current substance use (i.e., use in the
past 30 days); the difference between men with physical health conditions and other men at this
time was primarily due to lower alcohol intoxication and marijuana use. This decreased level of
postprison use occurred despite the fact that men with physical health conditions had reported
similar levels of preprison abuse and more lifetime years of substance use compared to other
men. Eight to ten months after prison, the portion reporting current substance use had risen to 3
in 10 but was still lower than the rate among other men; at this time point, the rate of alcohol
intoxication remained lower than other men but illegal drug use (most often marijuana and
cocaine use) was similar, reported by one-quarter. Men with physical health conditions were also
more likely than other men to have participated in substance abuse treatment programs and selfhelp groups like Alcoholics Anonymous (AA) and Narcotics Anonymous (NA), both during and
after prison (Figure 8). Moreover, men with physical health conditions had spent more
postrelease months in some type of recovery program relative to other men.
By contrast, women with physical health conditions were no different from other women
with respect to using drugs or getting intoxicated after prison, with about one-third reporting
current use (i.e., in the past 30 days) at each of the postrelease interviews. Similar to other
women, the most common substances used, in order, were cocaine, marijuana, and alcohol.
Women with physical health conditions had been more likely to participate in formal substance
abuse treatment programs in prison, though overall receipt of treatment services (including AA
and NA) was similar to other women. After release, they were more likely to report seeking help
for their substance abuse problems, yet their likelihood of participating in treatment services
(including self-help groups like AA and NA) was no different from other women. Nevertheless,
Health and Prisoner Reentry

those women with physical health conditions who received treatment services reported spending
more postrelease months in treatment than other women.
Figure 7. Percentage of respondents reporting substance use, by physical health status and gender
Men with physical health conditions were less likely than other men to engage in postrelease substance use, despite
similar rates of preprison use. Among women, however, postrelease use did not differ by physical health status.
100%

100%

Men

80%

60%

60%
no physical
illness

40%

Women

40%

all women
(no significant differences)

physical
illness

20%

0%
Before prison

2-3 months
postrelease**

8-10 months
postrelease**

Before prison

2-3 months
postrelease

8-10 months
postrelease

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant. Substance use includes any drug use or alcohol intoxication; preprison measures reflect use during the
six months before this incarceration and postrelease measures reflect use within 30 days of the study interview.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 828 men and 250
women. Two to three month postrelease measures are based on 657 men and 193. Eight to ten month
postrelease measures are based on 537 men and 150 women.

Figure 8. Percentage who participated in substance abuse treatment services, including Alcoholics or
Narcotics Anonymous, by physical health status and gender
Among men, those with physical health conditions received substance abuse treatment services at higher rates before
and after prison. However, there were no significant differences among women by physical health status.
Men

100%

80%

60%

40%

Women

40%
physical
illness

20%

no
physical
illness

all women
(no significant differences)

20%

0%
During prison**

2-3 months
postrelease**

8-10 months
postrelease**

During prison

2-3 months
postrelease

8-10 months
postrelease

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant. Substance abuse treatment services include formal drug and alcohol treatment programs as well as
self-help groups like Alcoholics Anonymous or Narcotics Anonymous; postrelease statistics reflect participation in
substance abuse treatment services within 30 days of the study interview.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 824 men and 249
women. Two to three month postrelease measures are based on 657 men and 193 women. Eight to ten month
postrelease measures are based on 536 men and 149 women.

Health and Prisoner Reentry

Criminal involvement and physical health

Returning prisoners with physical health conditions had higher rates of criminal involvement
after prison, but the interplay between physical health, criminal activity, and criminal justice
system response differed by gender (Figure 9). Men with physical health conditions were no
different from other men in terms of reporting criminal behavior after release, yet they were
reincarcerated at higher rates. One year after release, one-fifth of men with physical health
conditions were back in state prisons, a rate higher than other returning men (15 percent). While
the majority of reincarcerations were for new crimes, men with physical health conditions were
more likely to be back in prison for parole violations (9 percent compared to 4 percent of other
men). However, this higher reincarceration rate was not reflective of greater engagement in
criminal activity. Close to 4 in 10 reported some criminal activity, arrests, or violations of
supervision conditions in the first eight to ten months after release, but this was no different from
other men. Additionally, the crime types reported by men with physical health conditions were
similar, and they were no more or less likely than other men to have engaged in illegal activities
for income.
Figure 9. Postrelease criminal involvement and criminal justice system response, by physical health
status and gender
Men with physical health conditions were reincarcerated at higher rates than other men, despite reporting lower levels
of substance use and similar levels of criminal activity. By contrast, women with physical health conditions reported
more criminal activity without experiencing significantly higher reincarceration.
100%

100%

Men

Women

80%

60%

physical illness
no physical illness
39%

40%

36%

40%

60%

40%

29%

53%
39%

physical illness
no physical illness
38%

28%
23%

20%

15%

17%

20%

0%
Postrelease
substance use**

Criminal behavior

Reincarcerated within
1 year*

Postrelease
substance use

Criminal behavior*

Reincarcerated within
1 year

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10. A double asterisk (**) denotes differences with a pvalue ≤ 0.05.
Source: Returning Home original data collection in Ohio and Texas. Postrelease substance use reflects any drug use or
alcohol intoxication within 30 days of the eight to ten month postrelease interview and is based on 537 men and
150 women. Criminal behavior includes any self-reported crimes, arrests, or parole violations during the first eight
to ten months after release and is based on 539 men and 150 women. Reincarceration was measured using
official records 12 months after release and is based on 804 men and 252 women.

Women with physical health conditions had the opposite experience: they were more likely
than other women to report criminal behavior, placing them at risk for reincarceration, but they
were not returned to prison at significantly higher rates. We also examined self-reported time in
jail after release and found no significant differences by physical health status. Over one-half had
“recidivated” in the first eight to ten months after prison, as measured by a combination of selfHealth and Prisoner Reentry

reported crimes, arrests, and supervision violations; among other women, this figure was 38
percent. However, the reincarceration rate for women with physical health conditions, at 23
percent, was not significantly higher than the rate observed among other women. While the
majority of these reincarcerations were for new crimes, women with physical health conditions
were more likely than other women to have been reincarcerated for a parole violation. Men and
women were similar in this respect: those with physical health conditions were more likely than
others to have been reincarcerated for parole violations even though similar shares had been
under parole supervision.
CONCLUSIONS ABOUT PHYSICAL HEALTH AND REENTRY

To sum up, one-half of men and two-thirds of women reported having chronic physical health
conditions requiring long term management, but available treatment did not match the level of
need. The majority received some treatment for these conditions during prison, but about onethird of men and one-quarter of women with physical conditions did not. Eight to ten months
after release, the majority of respondents with physical health conditions had utilized some
health care services in the community, but treatment rates for specific health conditions declined,
suggesting that they received acute, episodic care and that the management of chronic health
conditions deteriorated. Eight to ten months after release, respondents were more apt to describe
their health as fair or poor.
Examining the reentry experiences of respondents with physical health conditions against
this backdrop of declining health, we found several ways in which employment, substance use,
and recidivism outcomes differed from healthier respondents. At the same time, there were
gender differences in how physical health mediated reentry experiences, as physical health
appeared to have a greater influence on reentry outcomes among men than women.
Returning prisoners with physical health conditions had poorer employment outcomes. Both
men and women were less likely to have found any employment in the eight to ten months after
prison release. Although they had found employment in the first two to three months at rates
similar to other returning prisoners, employment rates were comparatively lower as time elapsed.
Compared to other returning prisoners, men and women with physical health conditions were
more likely to report that health problems interfered with their ability to work and they were
more dependent on disability assistance. Employment and income difficulties were more
pronounced among men. Eight to ten months after release, men with physical health conditions
were less likely than other men to support themselves through employment and more likely to
depend on public assistance and disability programs. Women with physical health conditions,
however, were not much different from other women: they were more likely to receive disability
payments but their sources of financial support were otherwise similar to other returning women,
with family and friends being their predominant source of income.
Correlations between physical health and substance use after release were gender specific.
Men with physical health conditions were less likely than other men to engage in substance use
after prison. This was driven by a difference in postrelease alcohol intoxication that was
Health and Prisoner Reentry

observed only among men. Men with physical health conditions were also more likely to have
participated in substance abuse treatment services (including AA or NA) both during and after
prison. By contrast, women with physical health conditions were not much different from other
women with regard to postrelease substance use or treatment.
Physical health was also related to recidivism, but in different ways for men and for women.
Men with physical health conditions were no different from other men in terms of engaging in
criminal behavior after release, yet they were returned to prison at higher rates. By contrast,
women with physical health conditions were more likely to engage in criminal activity but no
different in terms of reincarceration to either state prisons or local jails. At the same time,
regardless of gender, respondents with physical health conditions were more likely to be
reincarcerated for parole violations. This dynamic between physical health status, criminality,
and the criminal justice system response—particularly parole violations—calls for further
examination.
Interestingly, physical health status did not have much bearing on housing or family
outcomes. While respondents with physical health conditions could benefit from housing
assistance, their housing outcomes were not much different from others. Eight to ten months
after release, they did not experience homelessness at greater rates, even though they had been
less likely to have made housing arrangements before release and were more likely to shift
residences postrelease. This positive outcome is intrinsically linked to supportive family
relationships, since family members served as the predominant source of postrelease housing for
returning prisoners. Men and women with physical health conditions reported supportive family
relationships, and were as likely as others to have lived with family members in the first eight to
ten months after release.

Health and Prisoner Reentry

4 Mental Health and Reentry

MENTAL HEALTH STATUS AND TREATMENT

When we surveyed prisoners in the month before release, 15 percent of men and 35 percent of
women reported having been diagnosed with a mental health condition. This is a conservative
estimate of mental illness in correctional populations. First, it is a self-reported measure; given
that prisoners are not universally screened for such conditions, many respondents may be
unaware of having a mental health condition. Moreover, these data were collected in prison and,
while respondents were assured of the confidentiality of their responses, some may have been
reluctant to divulge their mental health status: when we asked this question again after release,
the portion reporting mental health conditions increased by about one-third.
Table 4. Prevalence estimates of mental illness among returning prisoners
Male Returning
Home participants

Female Returning
Home participants

U.S. correctional
population

Self-reported (during prison)**

14.9%

34.9%

24.3%

Self-reported (postrelease)**

18.5%

45.1%

–

Depression**

29.7%

56.4%

7.9–15.2%

Posttraumatic stress disorder**

16.1%

31.2%

4.0–8.3%

Any self-reported or estimated mental illness**

40.6%

70.8%

56.2%

Potentially undiagnosed mental illness

22.1%

25.7%

31.9%

Note: T-tests were conducted to identify statistically significant gender differences. A single asterisk (*) denotes
differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with a p-value ≤ 0.05.
Significance testing of gender differences was restricted to Texas participants to control for state effects; these
statistics are available from the authors upon request.
Sources: Returning Home estimates are based on original data collection in Ohio and Texas. In-prison measures are
based on 831 men and 252 women and postrelease measures are based on 665 men and 202 women who were
interviewed two to three months after release. Respondents self-reported whether a doctor or nurse had diagnosed
them with depression or other mental illness. Depression and PTSD were estimated using symptom scales that
were incorporated into the study interview (Radloff 1991; Foa et. al. 1993). “Any self-reported or estimated mental
illness” is comprised of those who reported a mental health diagnosis or reported symptomatology consistent with
the diagnostic criteria for depression or PTSD. The estimate of “potentially undiagnosed mental illness” is based
on those who reported symptomatology meeting the criteria for depression or PTSD but did not report a mental
health diagnosis. Estimates of mental illness in the U.S. correctional population are from James and Glaze, 2006.
Estimates of major depression and PTSD in the U.S. correctional population are from NCCHC, 2002. Dashes
indicate that comparable national estimates were not available.

Given these issues, we developed a range of estimates of mental illness based on multiple
measures before and after release from prison (Table 4). Two to three months after release, 19
percent of men and 45 percent of women reported having been diagnosed with a mental health
condition. During the postrelease interviews, we also administered screening tools for depression

Health and Prisoner Reentry

and posttraumatic stress disorder (PTSD).3 We estimate that 30 percent of men and 56 percent of
women had symptoms consistent with depression and that 16 percent of men and 31 percent of
women had symptoms consistent with PTSD. Looking across self-reports and the results of the
screening tools, we estimate that 41 percent of men and 71 percent of women were likely to have
a mental illness. Comparing self-reports to the screening results, we estimate that one-quarter of
returning prisoners are likely to have an undiagnosed mental health condition. This is consistent
with other research examining mental illness in correctional populations (James and Glaze
2006).
Additionally, some returning prisoners reported mental health symptoms that indicate an
immediate need for help and have implications for public safety. Six percent of men and 14
percent of women reported having had hallucinations within 30 days of the 8 to 10 month study
interview. Roughly 1 in 10 men and women said they experienced trouble controlling violent
behavior. Three percent of men and significantly more women (12 percent) reported having
suicidal thoughts, with 2 percent of men and 1 percent of women having attempted suicide since
their release.
Analyses for this report focus on the most conservative measure, the 15 percent of men and
35 percent of women who reported having a mental health condition during prison. Using this
definition of mental illness allows us to estimate mental health treatment coverage and gauge
continuity of care from prison to the community. It also allows us to look longitudinally at the
experiences of those with mental illness as they leave prison and transition back to the
community.
Men and women who reported having a mental health condition during prison typically
reported having other physical and substance abuse problems as well (Figure 2, shown
previously). Four out of ten respondents with mental health conditions reported being in fair or
poor health and they were more likely than others to give a negative assessment of their health.
Among those with mental health conditions, roughly two-thirds also had physical health
conditions and 7 out of 10 had substance abuse problems. Over 4 in 10 individuals with mental
health conditions reported having physical, mental, and substance abuse problems
simultaneously. Interestingly, respondents with mental health conditions were more likely than
others to report a history of alcohol intoxication and women, in particular, reported recent
preprison alcohol abuse. Men with mental health conditions were also more likely to have
abused cocaine before this prison term.
About 6 in 10 men and women with mental health conditions received mental health
treatment during prison, meaning that about 4 in 10 did not. Mental health treatment rates
declined to roughly 50 percent immediately after prison and stayed relatively constant eight to
ten months after release (Table 5). Continuity of prescription medication also suffered, especially
3

Depression was estimated using the Center for Epidemiologic Studies depression scale, with scores of 16 and
above indicating a high likelihood of clinical depression (Radloff 1991). Posttraumatic stress disorder was estimated
using an adaptation of the 17-item PTSD Symptom Scale; scale items and scoring correspond to the DSM-III-R
diagnostic criteria for PTSD (Foa 1993).
Health and Prisoner Reentry

among women. Most respondents who used prescription medication regularly during prison (for
either physical or mental health reasons) received a supply of medicine at release, yet medication
use dropped substantially afterwards. Two to three months after release, 74 percent of men and
60 percent of women who used medications in prison reported they were still using the same
type of medication. Eight to ten months after release, just 59 percent of men and 40 percent of
women were using the same type of medication. When asked about the reasons for this decline,
the men and women who stopped using prescription medications after release most often cited
“other” reasons and, secondarily, that they themselves decided they no longer needed the
medication. Women with mental health conditions, but not men, were similarly likely to report
cost as a reason for stopping medication. Additionally, both men and women with mental health
conditions reported that a doctor in the community determined that the medication was no longer
necessary.
Table 5. Percentage of respondents with mental illness who received mental health treatment, by gender
Men

Women

During prison

60.5%

56.8%

2-3 months after release

46.3%

55.0%

8-10 months after release

52.8%

42.4%

Note: T-tests were conducted to identify statistically significant gender differences; this analysis was restricted to Texas
participants to control for state effects. Mental health treatment rates were not significantly different by gender.
Source: Returning Home original data collection in Ohio and Texas. Mental health treatment rates are based on the
number of respondents who reported having a mental health condition at each study time point: 124 men and 88
women during prison, 123 men and 91 women two to three months postrelease, and 108 men and 66 women
eight to ten months postrelease.

A lack of health insurance was likely a barrier to continuous care for returning prisoners
with mental health conditions. Although respondents with mental health conditions were more
likely than others to have insurance shortly after release, the majority did not. About 3 in 10 men
and 4 in 10 women had some type of health insurance two to three months after prison. Eight to
ten months after prison, coverage rates for men and women with mental health conditions were
around 40 percent. The primary sources of health insurance for people with mental health
conditions were Medicaid (about 20 percent), private insurance (6 percent) and, among men,
veterans’ health benefits (6 percent). Respondents also reported other sources of coverage,
including local health care subsidy programs.
Despite a general lack of health insurance, most returning prisoners with mental health
conditions accessed health services after prison; in fact, they were more likely than others to
have done so, particularly in the early postrelease period. One-half of mentally ill men and twothirds of mentally ill women had received some health care in the first two to three months after
prison. Among mentally ill respondents, one-fifth of men and one-third of women had used
emergency room services in the first two to three months; additionally, one-tenth of men and
one-fifth of women had been hospitalized. Eight to ten months after prison, nearly 8 in 10
returning prisoners with mental health conditions had received some health care. By this time,
more than one-third had used emergency room services. Additionally one-fifth of men and twofifths of women had been hospitalized.

Health and Prisoner Reentry

REENTRY EXPERIENCES OF PRISONERS WITH MENTAL HEALTH CONDITIONS

There are many reasons to expect mentally ill persons to have a more difficult reentry experience
than others. Respondents with mental health conditions reported multiple health conditions and
were more likely than others to feel that they were in fair or poor health. Despite high rates of
health care utilization in the first eight to ten months after release, just one-half reported
receiving treatment for their mental health conditions. This suggests that respondents with
mental health conditions managed to receive episodic, acute care for physical or mental problems
but were not able to maintain continuous treatment for their mental health conditions. After
release, about 4 in 10 respondents with mental health conditions felt that health problems
affected their ability to work or perform other activities. In the following sections we examine
how people with mental health conditions navigated the reentry process and in what ways they
had distinctly different outcomes.
Housing and mental health

Returning men and women with mental health conditions had more housing difficulties
compared to other returning prisoners. Though large shares relied on family members for
housing, at times they were less likely than other returning prisoners to have lived with family
members and more likely to have experienced homelessness after prison (Figures 10 and 11).
These difficulties are, perhaps, predictable since men and women with mental health conditions
reported high rates of preprison homelessness. When we surveyed them in the month before
Figure 10. Percentage of returning prisoners who ever lived with family members, by mental health status
and gender
Returning prisoners with mental illness were less likely to have lived with family members, particularly in the first few
months after prison release.
100%

100%

Men

80%

no mental illness

60%

mental illness

Women

80%
no mental
illness

60%
mental
illness

40%

20%

0%
Before prison

1st night
postrelease*

2-3 months
postrelease**

8-10 months
postrelease

Before prison

1st night
postrelease

2-3 months
postrelease**

8-10 months
postrelease*

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 824 men and 250
women. First night out measures are based on 657 men and 193 women. Two to three month postrelease
measures are based on 658 men and 193 women. Eight to ten month postrelease measures are based on 539
men and 150 women.

Health and Prisoner Reentry

release, they had been less likely than others to have made arrangements for postrelease housing
and they were less likely to expect to live with family members. Although rates of homelessness
were lower postrelease than they had been before prison, those with mental health conditions
were more likely than others to report that they had trouble keeping housing. They were also
more likely than others to report living with other former prisoners and current substance
abusers, individuals that could potentially jeopardize their reentry success.
Figure 11. Rates of homelessness, by mental health status and gender
Returning prisoners with mental health conditions reported higher rates of homelessness than others. Mentally ill
women experienced disproportionate homelessness early in the postrelease period, whereas this occurred later on for
mentally ill men.
100%

100%

Men

80%

60%

40%

mental
illness

20%

Women

mental illness

20%

none

0%
Before prison

1st night
postrelease

2-3 months
postrelease

8-10 months
postrelease

Before prison

1st night
postrelease

2-3 months
postrelease

8-10 months
postrelease

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 824 men and 250
women. First night out measures are based on 657 men and 193 women. Two to three month postrelease
measures are based on 658 men and 193 women. Eight to ten month postrelease measures are based on 539
men and 150 women.

Although mentally ill men and women both experienced housing difficulties, the timing of
their problems differed. Mentally ill men’s housing problems were more pronounced in the later
part of the postrelease period. Although they had been less likely to live with family in the first
two to three months of release, their rates of homelessness at that time were similar to other men.
Eight to ten months after release, however, men with mental health conditions reported greater
residential mobility than other men and were more likely to have been homeless. At this point in
the postrelease period, they had been as likely as other men to have lived with family for some
time, but they had spent fewer postrelease months living with family members. By contrast,
women with mental health conditions had more acute housing problems in the first two to three
months after release, reporting more residential mobility and homelessness (14 percent) relative
to other women. Eight to ten months after release, their rates of homelessness were not any
different from other women’s since homelessness among women generally increased over time.
Though they often lived with family members, women with mental health conditions had been
less likely than other women to have lived with family members beyond the first night after
release.

Health and Prisoner Reentry

Employment and mental health

Men and women with mental health conditions had poorer employment outcomes after release
compared to other returning prisoners, despite having had similar employment histories before
this prison term. We found that men and women with mental health conditions had been less
likely to find any employment, worked for fewer postrelease months, and were less likely to
have current employment at the each of the postrelease interviews. These poor outcomes were
observed early in the postrelease period and persisted through the 8 to 10 month follow-up
interview (Figure 12). Just 36 percent of men and 26 percent of women with mental health
conditions had found any employment within the first 2 to 3 months out of prison. Eight to ten
months after release the shares reporting any postrelease employment were 59 percent among
mentally ill men and 49 percent among mentally ill women. These rates of employment were
significantly lower than the rates reported by other returning prisoners. Moreover, 8 to 10 months
postrelease, returning prisoners generally reported employment rates in line with or exceeding
their preprison employment levels; those with mental health conditions, however, had not caught
up to their levels of preprison employment.
Figure 12. Percentage of respondents who were ever employed, by mental health status and gender
Returning prisoners with mental illness had less success in finding employment after release compared to others,
despite having had similar employment levels before this incarceration.
100%

100%

Men

80%

no mental
illness

80%

60%

mental
illness

60%

Women

no mental
illness

mental
illness

40%

20%

0%
Before prison

2-3 months
postrelease**

8-10 months
postrelease**

Before prison

2-3 months
postrelease**

8-10 months
postrelease**

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 808 men and 248
women. Two to three month postrelease measures are based on 657 men and 190 women. Eight to ten month
postrelease measures are based on 536 men and 148 women.

In keeping with their lowered employment, men and women with mental health conditions
were less likely than others to support themselves financially through work and more likely to
rely on other sources for income (Figure 13). The majority of men and women with mental
health conditions reported receiving some income from family members and friends. Eight to ten
months after release, just 3 in 10 men and 2 in 10 women with mental health conditions were
supporting themselves through legal employment, rates that were significantly lower than other

Health and Prisoner Reentry

returning prisoners. This was offset, somewhat, by the fact that men and women with mental
health conditions were more likely than others to receive Social Security disability payments.
There were, however, gender differences beyond these broad similarities. Although family
members and friends were the most common source of financial support reported by mentally ill
men and women eight to ten months after prison, the likelihood of receiving family support
differed by gender. Men with mental health conditions had a similar or greater likelihood of
receiving financial support from family members eight to ten months after release (depending on
the specific measure used) but women with mental health conditions were less likely than other
returning women to receive financial support from family members (p=.13 and this finding is
consistent with other measures of family financial support not shown in Figure 13). A similar
pattern was observed with regard to public assistance programs other than disability. Men with
mental health conditions were more likely than other men to receive benefits such as food
stamps. Women with mental health conditions, however, were no different from other women in
terms of receiving assistance. In sum, men with mental health conditions had a similar or greater
likelihood of receiving various forms of financial assistance relative to other men, while women
with mental health conditions typically had a similar or lower likelihood of receiving these
supports compared to other women.
Figure 13. Sources of income eight to ten months after prison release, by mental health status and gender
Returning prisoners with mental illness were less likely to report income from a job and more likely to rely on disability
assistance. Friends and family were the most common source of financial support.
Men

Women

28%

Legal employment**

18%

Legal employment**

35%

53%
32%
32%

"Under the table" work

57%

Family and friends**

18%
24%

"Under the table" work

53%

Family and friends

65%

41%
15%

SSI/SSDI disability**

31%

SSI/SSDI disability**

4%
mental illness
no mental illness

30%

Public assistance**

16%
6%
6%

Illegal activities

18%
20%

Public assistance

18%
14%

Illegal activities

20%

40%

60%

80%

100%

mental illness
no mental illness

20%

40%

60%

80%

100%

Family support and mental health

The majority of men and women with mental health conditions received some tangible assistance
from family members, yet respondents with mental health conditions reported receiving less
tangible and emotional support from their families when compared to other returning prisoners.
As discussed above, they were often less likely to have lived with family members since release

Health and Prisoner Reentry

(Figure 10, shown previously) and less likely to have received financial support from family
members or friends (Figure 13, shown previously).
There were gender differences in the timing of family assistance. Men with mental health
conditions were less likely than other men to receive tangible family support early in the
postrelease period, but levels of support increased over time and became similar to other men by
the 8 to 10 month follow-up interview. Men with mental health conditions were less likely than
other men to have lived with family members during the first 2 to 3 months after prison, even on
the first night out. Similarly, they were less likely to have received financial support from family
members in the first two to three months. However, family support increased later in the
postrelease period so that eight to ten months after prison they were as likely as other men to
have lived with family members or received financial support. Nonetheless, men with mental
health conditions perceived lower levels of emotional support throughout the postrelease period
and were less likely than other men to feel that they could count on family members for tangible
support.
Women with mental health conditions were less likely than other women to have lived with
family members throughout the postrelease period; while mentally ill men became more likely to
live with family after the first few months, this did not happen for women. With regard to
financial support, mentally ill women had more support early in the postrelease period, but this
declined by the time of the 8 to10 month interview. Two to three months after prison, women
with mental health conditions were as likely as other women to have received financial
assistance from family members, but, eight to ten months after release, they were less likely to
have received such assistance.
Differential levels of family support may be related to a number of factors. In some cases,
the family’s capacity to provide assistance is diminished: respondents with mental health
conditions were more likely than other returning prisoners to report familial problems with
substance abuse. Domestic violence may be another reason: 15 percent of men and 40 percent of
women with mental health conditions reported having been victimized by a family member
before this incarceration. Women with mental health conditions were also more likely to have
perpetrated violence against a family member before this incarceration.
Substance use and mental health

Men and women with mental health conditions reported patterns of postrelease drug and alcohol
use that were generally similar to other returning prisoners, despite a greater history of preprison
alcohol abuse. Slightly over one-third reported current drug use or alcohol intoxication (i.e., in
the past 30 days) at the time of the eight to ten month postrelease interview. Among men, levels
of substance use increased over the postrelease period, with about one-quarter reporting drug use
or alcohol intoxication at the time of the two to three month postrelease interview. Among
women, levels of substance use remained constant over the postrelease period, with about onethird reporting use or alcohol intoxication at the time of the two to three month postrelease
interview. Like other returning prisoners, women with mental health conditions were more likely
than men to have used illegal drugs after release.
Health and Prisoner Reentry

These similar outcomes, despite greater preprison alcohol abuse (and, among men, greater
preprison cocaine use as well) may be related to increased participation in substance abuse
treatment services during prison. Men and women with mental health conditions were more
likely than other prisoners to have participated in self-help programs like Alcoholics Anonymous
(AA) or Narcotics Anonymous (NA) during prison. Figure 14 illustrates that roughly one-half of
men and women with mental health conditions had participated in some substance abuse
treatment services (including substance abuse treatment programs, AA or NA) while
incarcerated. After release, as with other returning prisoners, there was a substantial drop in the
share receiving substance abuse treatment services. Two to three months after prison, those with
mental health conditions were statistically no different from other returning prisoners, with
roughly 3 in 10 receiving such services at the time of the interview. Eight to ten months after
release, however, there is some evidence to suggest that men and women with mental health
conditions were more likely to receive substance abuse treatment services than other returning
prisoners. Men with mental health conditions were more likely than other men to report current
participation (i.e., within the past 30 days) in either substance abuse treatment or self-help groups
at the time of the second study interview. Women with mental health conditions reported
spending more postrelease months in substance abuse treatment services over the course of the
postrelease period, though there was no difference from other women in current receipt of
substance abuse treatment services at the time of the 8 to 10 month postrelease interview.
Figure 14. Percentage who participated in substance abuse treatment services, including Alcoholics or
Narcotics Anonymous, by mental health status and gender
Returning prisoners with mental health conditions had been more likely to participate in substance abuse treatment
services during prison, particularly self-help groups like Alcoholics or Narcotics Anonymous.
100%

100%

Men

80%

60%

40%

mental
illness
no mental
illness

20%

Women

mental
illness

40%
no mental
illness

20%

0%
During prison**

2-3 months
postrelease

8-10 months
postrelease*

During prison**

2-3 months
postrelease

8-10 months
postrelease

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant. Substance abuse treatment services include formal drug and alcohol treatment programs as well as
self-help groups like Alcoholics Anonymous or Narcotics Anonymous; postrelease measures reflect participation in
substance abuse treatment services within 30 days of the study interview.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 824 men and 249
women. Two to three month postrelease measures are based on 657 men and 193 women. Eight to ten month
postrelease measures are based on 536 men and 149 women.

Health and Prisoner Reentry

Criminal involvement and mental health

Men and women with mental health conditions reported more postrelease criminal behavior than
other returning prisoners (Figure 15), but the criminal justice system response differed somewhat
by gender. Men with mental health conditions were not arrested, jailed, or reincarcerated in state
prison at significantly different rates than other men. Women with mental health conditions,
however, were more likely to be arrested than other women in the first two to three months after
release, yet they were no more or less likely to be returned to incarceration in a state prison.
Interestingly, men and women with mental health conditions were less likely than others to be
reincarcerated for parole violations, despite similar odds of being under parole supervision and
similar reports of violating their supervision conditions.
Figure 15. Postrelease criminal activity and justice system response, by mental health status and gender
Returning prisoners with mental illness, particularly men, reported more criminal behavior than others, yet they were not
reincarcerated at different rates.
100%

100%

Men

80%

60%

mental illness
no mental illness

45%

40%

Women

35%

34%

36%

17%

53%

mental illness
no mental illnesss

45%

40%
17%

20%

60%

36%

35%
20%

21%

20%

0%
Postrelease
substance use

Criminal behavior*

Reincarcerated within
1 year

Postrelease
substance use

Criminal behavior

Reincarcerated within
1 year

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10. A double asterisk (**) denotes differences with a pvalue ≤ 0.05. Returning women with mental illness had reported significantly more criminal behavior than other
women two to three months after release.
Source: Returning Home original data collection in Ohio and Texas. Postrelease substance use reflects any drug use or
alcohol intoxication within 30 days of the eight to ten month postrelease interview and is based on 537 men and 150
women. Criminal behavior includes any self-reported crimes, arrests, or parole violations during the first eight to ten
months after release and is based on 539 men and 150 women. Reincarceration was measured using official records
12 months after release and is based on 804 men and 252 women.

Men with mental health conditions were more likely than other men to report committing
crime during the first eight to ten months out of prison. However, they were no more likely than
other men to have been arrested. One year after release, the rate of reincarceration for men with
mental health conditions, at 17 percent, was the same as that for other men. Interestingly, among
those on parole, men with mental health conditions did not report violating supervision
conditions any more or less than other men. However, their rate of reincarceration for parole
violations was less than other men.
Women with mental health conditions reported more criminal behavior than other women in
the first two to three months after release. Eight to ten months after release, self-reported crimes

Health and Prisoner Reentry

were higher than other women, but the difference was not statistically significant (p=.17). Unlike
men, however, women with mental health conditions were arrested at higher rates than other
women. The data also suggest that women with mental health conditions were somewhat more
likely to spend at least one postrelease month in jail (p=.20). However, higher arrest rates did not
translate into higher rates of return to prison. One year after release, 20 percent of women with
mental health conditions were reincarcerated in state prisons, a rate similar to other women.
Interestingly, women with mental health conditions were less likely than other women to be
reincarcerated for parole violations even though they were no different from other women in
reporting supervision violations.
CONCLUSIONS ABOUT MENTAL HEALTH AND REENTRY

Returning prisoners with mental health conditions faced many health and reentry challenges
upon release. Respondents with mental health conditions were heavy users of health services
after release, but the data suggest that they received fragmented, episodic care for acute
problems. Eight to ten months after release, 8 in 10 respondents with mental health conditions
had received some health care in the community, but only one-half reported receiving treatment
for their mental health conditions. Respondents with mental health conditions were more likely
than others to have utilized emergency room services and to have been hospitalized.
Respondents with mental health conditions reported poorer housing and employment
outcomes relative to other returning prisoners. Men and women with mental health conditions
were more likely to have been homeless after prison, though homelessness rates 8 to 10 months
after prison were lower than the preprison rates reported by respondents. Additionally,
respondents with mental health conditions had been less likely to find employment after release,
often citing health problems as an impediment to work or other daily activities. Eight to ten
months after prison, they were less likely than others to receive income from legal employment
and more likely to receive disability support.
These poor housing and income outcomes reflect, to some extent, comparatively lower
levels of family support compared to other returning prisoners. Family members were key
sources of housing and income support for most returning prisoners. While the majority of those
with mental health conditions reported family support in these areas, they were less likely to
have lived with family members or received financial support from them. Moreover, those with
mental health conditions reported lower levels of emotional support from their families. We
hypothesize many reasons why, beyond the stigmatizing effect of mental illness, these
respondents experienced lower levels of family support. It may be that the family’s capacity to
provide assistance is diminished because of familial problems with substance use. Domestic
violence rates, both perpetration and victimization, were also higher relative to other returning
prisoners.
Family support for returning prisoners with mental health conditions varied over time and by
gender. In a sense, the family response to mental illness differed by the gender of the returning
prisoner. While both men and women with mental health conditions were less likely to have
Health and Prisoner Reentry

received family support in the early postrelease months, support for men increased over time so
that, eight to ten months after release, it was no different from the support reported by other men.
However, family support for women with mental health conditions remained lower than the
support given to other returning women. Considering that women generally had poorer outcomes
than men, regardless of their health conditions, the decreased level of support to women with
mental health conditions amounts to a double disadvantage due gender and health status.
Mentally ill respondents also reported higher levels of criminal involvement after prison, but
we did not observe consistent relationships between criminal behavior and criminal justice
system sanctions. Men and women with mental health conditions were more likely to report
committing crimes. Interestingly, this did not translate to higher reincarceration rates for either
new crimes or parole violations. However, there were some differences in the criminal justice
system response by gender. Among women, those with mental health conditions were more
likely to report arrests than others, though this increased arrest rate did not appear to influence
reincarceration rates. Among men, however, there were no significant differences in arrest or
reincarceration rates. It is not possible, with these data, to determine the appropriateness of the
justice system response to those with mental health conditions. More research is needed to
further our understanding about mental illness and recidivism.
Interestingly, mental health status did not have significant impacts on postrelease substance
use. Respondents with mental health conditions reported similar levels of postprison use, despite
a greater history of alcohol intoxication and, among men, cocaine use. Men and women with
mental health conditions had been more likely to participate in substance abuse treatment
services in prison, particularly groups like AA or NA. There is some evidence to suggest that
their participation in treatment postrelease was also greater. Nonetheless, respondents with
mental health conditions reported aggregate levels of postrelease substance use that were
statistically no different from other returning prisoners.

Health and Prisoner Reentry

5 Substance Abuse and Reentry

SUBSTANCE ABUSE STATUS AND TREATMENT

Most returning prisoners had a history of substance use. Over 8 in 10 respondents had used drugs
or gotten intoxicated at least once in the six months leading up to their prison term. Illegal drug
use had been similarly common, reported by 75 percent of men and 83 percent of women. Figure
16 shows the types of substances used in the six months before prison and respondents’
frequency of use. Men most commonly reported alcohol intoxication (57 percent), followed
closely by marijuana (56 percent) and cocaine use (45 percent). Women reported cocaine use
most often (65 percent), followed by alcohol intoxication (55 percent) and marijuana use (53
percent).
Figure 16. Substance use in the six months prior to incarceration, by gender
Over 8 in 10 men and women reported some drug use or alcohol intoxication in the six months before this
incarceration. The majority used substances more often than once a week, a level of use that is consistent with abuse
and dependence.
Men

Women

Alcohol intoxication

Marijuana

Heroin

Cocaine

Cocaine
daily use
> weekly use
any use

Amphetamines
Ecstasy

Other drugs

Any of the above

Any of the above
0%

20%

40%

60%

80%

100%

daily use
> weekly use
any use

20%

40%

60%

80%

100%

Source: Returning Home original data collection in Ohio and Texas based on 835 men and 260 women.

The majority of returning prisoners can be characterized as substance abusers, since
relatively few reported casual use in the six months before prison. About two-thirds of all men
and women reported using drugs or getting intoxicated more often than once a week in the six
months preceding this prison term, including many (45 percent of men and 59 percent of women)
who reported daily use. This study defined substance abuse as drug use or alcohol intoxication
more often than once a week in the six months before this prison term; this is distinct from
substance use, which refers to use at any frequency. Substance use more than once a week was
significantly correlated with signs of abuse and dependence. Those who reported drug use or
alcohol intoxication more than once a week reported more adverse consequences resulting from
their use, including problems at work or school, arguments with friends and family, and driving
Health and Prisoner Reentry

while impaired. Substance use more than once a week was also correlated with signs of addiction
such as increased tolerance, greater than intended use, and an inability to stop using despite
negative consequences.
Figure 17. Percentage who participated in substance abuse treatment services, including Alcoholics or
Narcotics Anonymous, by substance abuse status and gender
Men with preprison substance abuse problems received treatment services at higher rates compared to other men.
Treatment rates among women, however, did not vary by their level of preprison use.
Men

100%

Women

100%

80%

60%

40%

40%
substance
abusers

20%

all women
(no significant differences)

20%

nonabusers

0%
During prison**

2-3 months
postrelease**

8-10 months
postrelease**

During prison

2-3 months
postrelease

8-10 months
postrelease

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant. Substance abuse treatment services include formal drug and alcohol treatment programs as well as
self-help groups like Alcoholics Anonymous or Narcotics Anonymous; postrelease measures reflect participation in
substance abuse treatment services within 30 days of the study interview.
Source: Returning Home original data collection in Ohio and Texas. In-prison measures are based on 828 men and 255
women. Two to three month postrelease measures are based on 664 men and 200 women. Eight to ten month
postrelease measures are based on 541 men and 154 women.

Despite the high prevalence of substance use and abuse before prison, relatively few
returning prisoners reported receiving substance abuse treatment services while incarcerated. In
this report, we use the term treatment services to describe formal substance abuse treatment
modalities, like therapeutic communities and pharmaceutical treatment regimens, as well as selfhelp groups like Alcoholics Anonymous (AA) or Narcotics Anonymous (NA). Roughly 4 in 10
of all returning prisoners participated in some treatment services during prison, though just onequarter of men and 14 percent of women reported participating in a formal drug or alcohol
treatment program while incarcerated. Men who reported substance abuse before prison were
more like to have received treatment services, but much need remained unmet: 29 percent had
participated in a formal substance abuse treatment program in prison. When AA or NA
participation is also taken into account, about one-half of men with a recent substance abuse
history had received some type of treatment services during prison (Figure 17). However, the
substance abuse treatment services provided to incarcerated women did not seem well matched
to the level of need. Women who had reported substance abuse before prison were not any more
or less likely to have participated in substance abuse treatment services than other women; about
4 in 10 received some services during prison, including 14 percent who participated in formal
treatment programs. Interestingly, women with physical health conditions had been more likely
to participate in formal drug and alcohol treatment programs during prison (16 percent) and

Health and Prisoner Reentry

women with mental health conditions had been more likely to participate in AA and NA-type
groups during prison (47 percent).
Participation in substance abuse treatment services dropped considerably after release from
prison. Just 3 in 10 of all returning men and women participated in either substance abuse
treatment programs or self-help groups immediately after release and participation rates declined
to 2 in 10 over the course of the first postrelease year, with fewer than 10 percent participating in
formal treatment programs. Men who had reported preprison substance abuse experienced a
similar drop in treatment services after prison release but continued to receive such services in
the community at higher levels than other returning prisoners: two to three months after prison,
nearly one-third received treatment services (down from one-half in prison) and eight to ten
months after prison this figure was one-quarter. However, as it had been during prison, women
with substance abuse problems were no different from other women in terms of receiving
treatment services in the community.
Figure 18. Percentage reporting any drug use or alcohol intoxication, by substance abuse status and
gender
Men and women with preprison substance abuse problems were more likely to use again after release.
100%

100%

Men

Women

80%

60%

60%
substance
abusers

40%

20%

nonabusers

substance
abusers

40%

20%

nonabusers

0%
Before prison**

2-3 months
postrelease**

8-10 months
postrelease**

Before prison**

2-3 months
postrelease**

8-10 months
postrelease**

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant. Substance use includes any drug use or alcohol intoxication; preprison measures reflect use during the
six months before this incarceration and postrelease measures reflect use within 30 days of the study interview.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 835 men and 260
women. Two to three month postrelease measures are based on 664 men and 200 women. Eight to ten month
postrelease measures are based on 542 men and 155 women.

Not surprisingly, respondents who had substance abuse problems before prison were more
likely than others to use again after release (Figure 18). Eight to ten months after release, roughly
4 in 10 respondents with substance abuse problems reported current use, defined as drug use or
alcohol intoxication within 30 days of the study interview. Nonetheless, roughly 2 in 10
nonabusers also reported drug use or alcohol intoxication after prison. Postrelease use among
men was initially lower: 2 to 3 months after release, about one-third of men with substance abuse
problems reported current use. However, postrelease use among women began soon after release
and remained constant. Current use levels among substance abusing women were nearly the
same 2 to 3 months after release as they were 8 to 10 months postrelease.

Health and Prisoner Reentry

Comparing Figures 17 and 18, shown previously, it may seem than substance abusing men
resumed substance use at greater rates postrelease relative to other men, despite having received
treatment services at higher levels, but it is important to remember that these figures present
aggregate statistics. Those substance abusing men who received treatment services in prison
were significantly less likely than their untreated counterparts to use again after release.
Interestingly, substance abusing women who received treatment services in prison reported rates
of postrelease use that were similar to other, untreated women; receipt of treatment services in
prison did, however, increase their rates of treatment participation after release.
REENTRY EXPERIENCES OF PRISONERS WITH SUBSTANCE ABUSE PROBLEMS

Individuals who reported substance abuse in the six months before prison had long histories of
lifetime use yet often returned to the community with unresolved substance abuse problems.
About one-half of men and 4 in 10 women reported receiving treatment services while
incarcerated and even fewer participated in community-based treatment services upon release.
Many returned to families and communities that also had problems with substance use, placing
them at increased risk of relapse and other negative outcomes. Both illegal drug use and alcohol
intoxication (because it is a violation of parole conditions) place individuals at risk of rearrest
and reincarceration. Additionally substance abuse may interfere with the ability to secure
housing, maintain employment, and reestablish family relationships. In the sections that follow,
we examine the process by which individuals who reported preprison substance abuse (i.e., use
more than once a week) faced these reentry challenges and we identify the ways in which their
experiences were significantly different from those reported by other returning prisoners.
Housing and substance abuse

Returning prisoners who reported preprison substance abuse—about two-thirds of both men and
women—experienced distinctly different housing challenges by gender. While respondents who
reported preprison substance abuse had different housing outcomes relative to other returning
prisoners, substance abuse did not affect outcomes in the same way for men and women (Figures
19 and 20). Compared to other men, those with substance abuse problems experienced greater
challenges early in the postrelease period but were also more likely to be able to rely on family
members for housing. However, women with substance abuse problems were less likely to rely
on family members for housing early on and had greater problems with homelessness compared
to other women. Despite these differences, both men and women with substance abuse problems
were more likely to live with others who would pose a risk to their reentry success: eight to ten
months after release, one-third of men and one-half of women with substance abuse problems
were living with other former prisoners and substance abusers.
Men with substance abuse problems reported higher rates of preprison homelessness and
were less likely than other men to have made housing arrangements when we surveyed them in
the month before release. Though the majority, like other men, stayed with family members on
their first night out, they reported changing residences more often than other men during the first
two to three months after release. Nevertheless, men with substance abuse problems were more
Health and Prisoner Reentry

Figure 19. Percentage of returning prisoners who ever lived with family members, by substance abuse
status and gender
Men with substance abuse problems were more likely to live with family members after release. By contrast, women
with substance abuse problems had been less likely to live with family members in the early postrelease period.
100%

100%

Men

80%

substance abuse

60%

no abuse

Women

80%
no abuse

60%

substance abuse

40%

20%

0%
Before prison

1st night
postrelease

2-3 months
postrelease**

8-10 months
postrelease**

Before prison*

1st night
postrelease

2-3 months
postrelease*

8-10 months
postrelease

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 828 men and 258
women. First night out measures are based on 664 men and 200 women. Two to three month postrelease
measures are based on 665 men and 200 women. Eight to ten month postrelease measures are based on 543
men and 155 women.

Figure 20. Rates of homelessness, by substance abuse status and gender
Women with substance abuse problems were more likely than other women to experience homelessness after release.
Men with substance abuse problems, however, did not report differential rates of homelessness.
100%

100%

Men

80%

60%

40%

20%

substance
abuse

20%
no
abuse

no
abuse

Before prison

Women

0%
1st night
postrelease

2-3 months
postrelease

8-10 months
postrelease

Before prison

1st night
postrelease

2-3 months
postrelease

8-10 months
postrelease

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 828 men and 258
women. First night out measures are based on 664 men and 200 women. Two to three month postrelease
measures are based on 665 men and 200 women. Eight to ten month postrelease measures are based on 543
men and 155 women.

likely than other men to have lived with family members both early (two to three months) and
later on (8 to 10 months) in the postrelease period. Correspondingly, they were less likely than
other men to have lived in their own homes or apartments. Rates of homelessness throughout the

Health and Prisoner Reentry

first 8 to 10 months after release were around 5 percent and were no different than other men.
Although men with substance abuse problems did not report greater residential mobility eight to
ten months after release, they were more likely than other men to report trouble keeping housing.
It is also noteworthy that men with substance abuse problems were more likely than other men to
report having “lived” in jail for at least one postrelease month (12 percent compared to 7 percent,
p=.11).4
By contrast, women with substance abuse problems were less likely than other women to
live with family members during the first few months after release. This is consistent with their
preprison experience, and women with substance abuse problems had been less likely to expect
that they would live with family after release. However, eight to ten months after release, over
half had lived with family members at some point, a rate similar to other women. Still, women
with substance abuse problems reported moving around more often than other women as well as
having trouble keeping housing. Homelessness was a particular problem for women with
substance abuse histories, with 2 in 10 having experienced homelessness in the first 8 to 10
months after release. A similar share of women with substance abuse problems, 2 in 10, reported
“living” in jail for at least one postrelease month, compared to 6 percent of other returning
women.
Employment and substance abuse

Substance abuse before prison was related to poorer employment outcomes after release, with
more pronounced impacts among women than men. Men with substance abuse problems were
similar to other men in terms of having found employment in the first eight to ten months after
release, but their success in maintaining employment was lower. Women with substance abuse
problems had more difficulty than other women in both finding and keeping employment.
Figure 21 shows the percentage of respondents reporting any employment before and after
this incarceration. Both men and women with substance abuse problems experienced greater
unemployment before prison, but men with substance abuse problems managed to find
postrelease employment at rates similar to other returning prisoners. Women with substance
abuse problems initially found postrelease employment at rates similar to other women, but
impediments to gaining employment emerged after the first two to three months. Eight to ten
months after release, women with substance abuse problems had been less likely to work for any
length of time than other returning women. Comparing the numbers in Figure 21 to the rates of
current employment shown in Figure 22 sheds light on the influence of substance abuse in
maintaining employment. Both men and women with substance abuse problems reported fewer
months of postrelease employment compared to others. Two to three months after release, the
share of respondents with substance abuse problems who were currently employed at the time of
the interview was similar to other returning prisoners. Eight to ten months after release, however,
their rates of current employment were significantly lower than others. Among women, the
share reporting current employment eight to ten months postrelease declined from what it had
4

Postrelease housing was measured using a calendar that asked respondents to choose one category that best
described their housing situation in each postrelease month. Jail was listed among the response options.
Health and Prisoner Reentry

been two to three months after release. This is a particularly troubling finding, as other
subgroups of returning prisoners reported either constant or increasing rates of employment as
time elapsed.
Figure 21. Percentage of respondents who were ever employed, by substance abuse status and gender
Men with substance abuse problems were as likely as other men to have found employment after prison, but women
with substance abuse problems had less success compared to other women.
100%

80%

100%

Men

Women

80%

no abuse

60%

substance
abuse

60%

40%

20%

substance
abuse

0%
Before prison*

2-3 months
postrelease

8-10 months
postrelease

Before prison**

2-3 months
postrelease

8-10 months
postrelease*

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 813 men and 255
women. Two to three month postrelease measures are based on 664 men and 197 women. Eight to ten month
postrelease measures are based on 540 men and 153 women.

Figure 22. Percentage of respondents reporting current employment at the time of the interview, by
substance abuse status and gender
Returning prisoners with substance abuse problems had more difficulty maintaining employment, as evidenced by
lower rates of current employment eight to ten months after release.
100%

100%

Men

80%

80%
59%

60%

40%

Women

51%
37%

38%

60%

53%

40%
30%

32%
25%

20%

0%
2-3 months postrelease
substance abuse

8-10 months postrelease*
no abuse

2-3 months postrelease
substance abuse

8-10 months postrelease**
no abuse

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10 and a double asterisk (**) denotes differences with
a p-value ≤ 0.05. The aggregate statistic is shown when differences by health status were not statistically
significant.
Source: Returning Home original data collection in Ohio and Texas. Preprison measures are based on 813 men and 255
women. Two to three month postrelease measures are based on 665 men and 200 women. Eight to ten month
postrelease measures are based on 541 men and 155 women.

Health and Prisoner Reentry

These differential employment outcomes are reflected in the way that men and women with
substance abuse problems supported themselves financially. Eight to ten months after release,
men with substance abuse problems were more likely to report income from illegal activities, but
they were otherwise not different from other men. Legal employment was the most common
source of income, followed by family and friends, and “under the table” work; smaller shares
reported receiving public assistance, disability payments, and illegal income. Women with
substance abuse problems, however, supported themselves quite differently from other women.
Like other women, they cited family and friends as the most common source of income, but
women with substance abuse problems were three times as likely to have received income from
illegal activities (19 percent and 6 percent, respectively). Moreover, they were significantly less
likely to have earned money through legal employment and less likely to have received benefits
like food stamps from public assistance programs. The one commonality is that both women and
men with substance abuse problems had been more likely to support themselves, at least
partially, through illegal activities, placing them at increased risk for arrest and reincarceration.
Family support and substance abuse

Respondents with preprison substance abuse problems typically received some tangible support
from family members during the first 8 to 10 months after release, but the interaction between
substance abuse and family support was different for men and for women. Among men, those
with a substance abuse problem often received more family support compared to others without
such a problem. Men with substance abuse problems were more likely to have lived with family
members throughout the postrelease period (Figure 19) and, eight to ten months after release,
they were also more likely to report having received financial support from family members
(Figure 23). Women with substance abuse problems, however, received less consistent support.
Figure 23. Financial support from family members, by substance abuse status and gender
Men with substance abuse problems were similarly or more likely than other men to receive financial support from
family members. By contrast, women with substance abuse problems were similarly or less likely than other women to
receive financial support from family members.
100%

80%

100%

Men
80%

81%

Women

85%

82%

76%

80%

78%
73%

66%

60%

40%

20%

0%
2-3 months postrelease
substance abuse

8-10 months postrelease**
no abuse

2-3 months postrelease**
substance abuse

8-10 months postrelease
no abuse

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10. A double asterisk (**) denotes differences with a pvalue ≤ 0.05.
Source: Returning Home original data collection in Ohio and Texas. Two to three month postrelease measures are based
on 665 men and 200 women. Eight to ten month postrelease measures are based on 543 men and 155 women.

Health and Prisoner Reentry

Eight to ten months after release, they were as likely as other women to have lived with family
members or received financial support, but they had received comparatively less support in
earlier months. Two to three months after release, they had been less likely than other women to
live with family members or receive financial assistance. Throughout the postrelease period,
women with substance abuse problems perceived lower levels of emotional support and were
less likely than other women to feel that they could rely on family members for tangible support.
Family support notwithstanding, the family members of respondents with substance abuse
histories posed particular risks to their reentry success. Perhaps not surprisingly, the family
members of respondents with substance abuse problems were more likely to have their own
problems with drugs and alcohol. Respondents with substance abuse problems also reported
higher levels of familial criminality compared to other respondents. Family violence is another
consideration. Men and women with substance abuse problems were more likely than others to
have perpetrated violence against their family members before this incarceration; after release,
however, men with substance abuse problems were more likely to report victimization by family
members. Negative family dynamics such as these may factor into returning prisoners’ decisions
to seek or accept family support.
Criminal involvement and substance abuse

Men and women with substance abuse problems before this prison term were more likely than
other returning prisoners to report criminal behavior after release and to be reincarcerated. This
held true across a variety of recidivism measures (Figure 24). One year after release, one-fifth of
men and one-quarter of women with substance abuse problems were back in state prisons
compared to 12 percent and 9 percent, respectively, of those without substance abuse problems.
Those with substance abuse problems also reported higher criminal activity in the first eight to
ten months postrelease, more arrests, and were more likely to have spent at least one month in
jail. Among those on parole, men with substance abuse problems were more likely to report
violating their release conditions; women with substance abuse problems had a similar likelihood
of violating any conditions compared to other women, but they violated more conditions. All
told, 4 in 10 men and over one-half of the women with substance abuse problems reported
committing crime, being arrested, or violating their parole conditions.
The specific crimes reported by respondents with substance abuse problems were most often
drug possession, drug selling, robbery, theft, and, among women, “other” offenses, which are
likely to include prostitution. Not surprisingly, respondents with preprison substance abuse
problems were more likely than others to have reported drug possession after release. They were
also more likely than other respondents to report that illegal activities were a source of
postrelease income. More often than other men, those with substance abuse problems engaged in
drug selling and burglary after release. Women with substance abuse problems, on the other
hand, were more likely than other women to engage in theft and “other” offenses after release.

Health and Prisoner Reentry

Figure 24. Postrelease criminal activity and justice system response, by substance abuse status and
gender
Returning prisoners with preprison substance abuse problems reported more criminal behavior after release and were
more likely to be reincarcerated.
100%

100%

Men

80%

60%
44%

substance abuse
no abuse

41%

40%

20%

Women

55%

20%
12%

substance abuse
no abuse

44%

40%

31%
18%

60%

20%

31%

26%

16%
9%

0%
Postrelease
substance use**

Criminal behavior** Reincarcerated within
1 year**

Postrelease
substance use**

Criminal behavior** Reincarcerated within
1 year**

Note: A single asterisk (*) denotes differences with a p-value ≤ 0.10. A double asterisk (**) denotes differences with a
p-value ≤ 0.05.
Source: Returning Home original data collection in Ohio and Texas. Postrelease substance use reflects any drug use or
alcohol intoxication within 30 days of the eight to ten month postrelease interview and is based on 542 men and
155 women. Criminal behavior includes any self-reported crimes, arrests, or parole violations during the first eight
to ten months after release and is based on 543 men and 155 women. Reincarceration was measured using
official records 12 months after release and is based on 810 men and 260 women.

CONCLUSIONS ABOUT SUBSTANCE ABUSE AND REENTRY

The majority of returning men and women had considerable problems related to drug use and
alcohol intoxication. Over 8 in 10 respondents reported drug use or alcohol intoxication in the six
months preceding this prison term, but very little of this was casual use. About two-thirds of all
men and women reported using drugs or getting intoxicated more often than once a week in the
six months before prison, including many (45 percent of men and 59 percent of women) who
reported daily use. Drug use or alcohol intoxication more than once a week was associated with
an increased number of social and interpersonal problems as well as signs of abuse and
dependence.
About 4 in 10 of all returning prisoners utilized some form of treatment services during
prison, including formal drug or alcohol treatment programs and self-help groups like Alcoholics
or Narcotics Anonymous, and participation in such services dropped considerably upon release.
Many individuals who used and abused drugs and alcohol before prison remained untreated and
the study’s findings raise questions about whether scarce treatment resources are allocated
appropriately. In-prison treatment services and linkages to community-based services seemed
better matched to needs in men’s facilities: men with substance abuse problems had been more
likely than other men to participate in treatment services both during and after prison, while
women with substance abuse problems reported participation rates that were no different from
other women. Interestingly, women with physical health conditions had been more likely to

Health and Prisoner Reentry

participate in formal treatment programs during prison and women with mental health conditions
had been more likely participate in AA or NA-type groups during prison. It is troubling that
substance abusing women were not more likely to have participated in substance abuse treatment
services, but it may be that women with co-occurring physical and mental health conditions were
given higher priority. Further research on the allocation of treatment resources to needs would
shed light on these findings.
Eight to ten months after release, over one-third of men and women had used drugs or gotten
intoxicated postrelease, and men and women with preprison substance abuse problems were
more likely to have used again after release. The finding that women with substance abuse
problems were more likely to use again after release is consistent with the fact that they did not
receive treatment services at higher rates relative to other women. Substance abusing men,
however, did receive treatment services at higher rates, both during and after prison, yet
aggregate rates of postrelease substance use were higher compared to other men. This is not
because substance abuse treatment services were ineffective at an individual level: those men
who participated in treatment services during prison were less likely to use again after release.
Rather, this finding suggests that current (low) levels of treatment capacity, and perhaps also the
treatment modalities available, were not sufficient to make a difference in population-level
outcomes.
Examining the role of substance abuse in the reentry process, we found that returning
prisoners who had preprison substance abuse problems had poorer housing, employment and
recidivism outcomes, with women often experiencing worse outcomes than men. Looking at
housing, for example, men with substance abuse problems experienced more housing uncertainty
and mobility, but their rates of homelessness were not different from other men. Women with
substance abuse problems, however, experienced more homelessness relative to other women.
This difference in housing outcomes by gender can be traced to differential levels of family
support. Men with substance abuse problems were more likely to have lived with family
members after prison, and this likely shielded them from homelessness. Women with substance
abuse problems, on the other hand, were often less likely to have lived with family members.
Similarly, returning prisoners with substance abuse problems reported working for fewer
postrelease months compared to others, but more negative impacts were observed among women
than men. Men with substance abuse problems were as likely as other men to have had some
employment postrelease, but they were less likely to have maintained employment. Compared to
other women, those with substance abuse problems had greater difficulties with both finding and
keeping employment, and were less likely to have earned income through legal employment.
Returning prisoners’ housing and financial outcomes were often shaped by the extent to
which family members helped them. Our findings suggest that the family response to those with
substance abuse problems differed by the gender of the returning prisoner. Substance abusing
men were more likely to receive support from family members compared to other men.
Substance abusing women, however, received similar or lower levels of support compared to

Health and Prisoner Reentry

other women. This amounts to a double disadvantage with respect to family assistance: women
generally had less supportive family relationships than men, regardless of their substance abuse
status, and women with substance abuse problems reported poorer outcomes relative to other
women. This resulted in large magnitudes of difference between women and men with substance
abuse problems; the shares living with family, for example, differed on the order of 20
percentage points. Reasons for differential family support may include the capacity and
willingness of family members to provide support, family members’ own problems with crime
and substance use, and domestic violence.
Substance abuse before prison was clearly linked to recidivism. We examined two
dimensions of criminal involvement—criminal behavior and sanctions by the justice system—
and found that men and women with substance abuse problems were significantly more likely to
be involved in postrelease criminal activity and more likely to be reincarcerated. Regardless of
gender, they were more likely to report committing new crimes, violating parole conditions, and
earning money through illegal activities. The criminal justice system response reflected this
increased criminal activity. Respondents with substance abuse problems were arrested and
reincarcerated at higher rates compared to other returning prisoners.

Health and Prisoner Reentry

6 Discussion and policy implications

The findings presented in this report demonstrate that most returning prisoners have chronic
health conditions requiring treatment or management. Eight out of ten men and 9 out of 10
women reported having at least one physical health, mental health, and substance abuse problem.
One-half of men and two-thirds of women reported physical health conditions. Fifteen percent of
men and over one-third of women reported mental health conditions, and the actual prevalence is
likely to be double these self-reported amounts. Furthermore, about two-thirds reported active
substance abuse, not simply substance use, in the six months before this incarceration. Given the
high prevalence of these conditions, it is not sufficient to think of returning prisoners with health
conditions as a special-needs population but, rather, as the norm.
While many returning prisoners had received treatment during prison for their health
conditions, sizeable shares had not. Respondents with physical health conditions reported the
highest rates of treatment, with about two-thirds of men and three-quarters of women receiving
treatment in prison. Roughly 6 in 10 respondents with mental health conditions received mental
health treatment in prison. The lowest treatment rates were reported by substance abusers; about
one-half of men and 4 in 10 women with substance abuse problems participated in Alcoholics
Anonymous (AA), Narcotics Anonymous (NA), or formal substance abuse treatment programs
during prison. To some extent these treatment rates reflect a scarcity of available treatment but
other barriers, including insufficient screening and stigmatization, may need to be addressed.
Nonetheless, treatment rates in prison, however incomplete, were better than the rates reported
postrelease. As Figure 25 illustrates, treatment rates for all types of health conditions declined
upon release and respondents increasingly rated their health as fair or poor. Interestingly,
returning prisoners with physical and mental health conditions were heavy consumers of
postrelease health services, which included significantly more emergency room visits and
hospitalizations; this suggests that many respondents received episodic care for acute problems
but that continuous treatment of specific health conditions suffered.
Returning prisoners adjusted to life in the community within this context of declining health.
This report establishes, empirically, that returning prisoners followed distinct reentry trajectories
in the first postrelease year according to their health status. Returning prisoners with health
problems typically had poorer outcomes across a range of reentry domains, including housing,
employment, family relationships, substance use, and recidivism. The degree to which health
conditions affected reentry outcomes varied by the type of health condition. Respondents with
physical health conditions, including those with co-occurring mental health or substance abuse
problems, were the least different from other, healthier respondents, reporting differential
difficulties in some but not all domains. At the other end of the spectrum, respondents with
mental health conditions reported poorer outcomes across most domains. In between were
respondents with substance abuse problems, who were more likely to relapse into substance use

Health and Prisoner Reentry

and recidivate, but otherwise had differential experiences by gender. Table 6 summarizes the key
reentry outcomes we examined and shows how these outcomes differ according to health status.
Figure 25. Treatment rates for physical, mental, and substance abuse conditions over time compared to
respondent assessments of their overall health, by gender
Respondents increasingly rated their health as fair or poor as treatment rates for physical, mental, and substance
abuse problems declined after prison release.
100%

100%

Men

80%

Women

80%
phys. trmt

60%

60%
m.h. trmt
phys. trmt

40%

20%

m.h. trmt

40%
fair or poor
health

s.a. trmt

fair or poor
health

s.a. trmt

20%

0%
During prison

2-3 months
postrelease

8-10 months
postrelease

During prison

2-3 months
postrelease

8-10 months
postrelease

Source: Returning Home original data collection in Ohio and Texas. Physical and mental health treatment rates are
based on the number of respondents reporting each condition at each study time point. Substance abuse
treatment rates are based on the portion who reported preprison use more often than once a week. Selfassessments of health were collected from respondents at each study time point. Sample sizes are available from
the authors upon request.

This research has demonstrated how the transition from prison to community differs for
individuals with health conditions, and the knowledge of these specific differences can be used
to target interventions to improve reentry outcomes. However, practitioners and policymakers
must first recognize the scope of the problem. This study took a cross-disciplinary look at a
representative sample of returning prisoners and found that nearly all had some type of physical
health, mental health, or substance abuse problem. Policymakers and practitioners would be well
served to adopt a new paradigm that recognizes health as a universal rather than a special needs
concern among returning prisoners.
Furthermore, it is important to recognize that returning prisoners frequently reported
multiple types of health issues. Roughly 4 in 10 men and 6 in 10 women reported a combination
of physical health, mental health, and substance abuse concerns. The majority of returning
prisoners with physical health conditions and substance abuse problems reported at least one
other type of condition and nearly all who reported mental health conditions had other, cooccurring conditions. Complete attention to a given returning prisoner’s health needs may often
require a cross-disciplinary approach.
Given that health conditions influence reentry outcomes, and that nearly all returning
prisoners have health issues, an assessment of health needs should be part of each individual’s
reentry planning process. Prison medical records are not sufficient to gauge health needs, as we
found that sizeable shares of returning prisoners who had health conditions did not utilize prison
health services while incarcerated (40 to 50 percent of men with physical or mental health

Health and Prisoner Reentry

conditions, for example, and the share of untreated substance abusers was higher still). A
prerelease health assessment could be as simple as a self-administered checklist to screen for
problems potentially requiring follow-up care or management. The Returning Home study
questionnaire, for example, asked respondents to check the physical and mental health conditions
they had been diagnosed with; potential substance abuse problems were assessed by asking about
the frequency of use before prison to minimize underreporting.
Table 6. Key reentry outcomes, by gender, with significant differences by health condition highlighted
Among respondents with—
Aggregate

Physical
illness

Mental
illness

Substance
abuse

Housing
Lived with family members, in the first 2-3 months
Lived with family members, in the first 8-10 months
Any homelessness, in the first 2-3 months
Any homelessness, in the first 8-10 months

Men

63%

52%

66%

Women

50%

40%

46%
73%

Men

70%

Women

56%

50%

Men

Women

11%

14%

10%

Men

13%

18%

Women
Employment and income
Any employment since release
Current employment, 2-3 months postrelease
Current employment, 8-10 months postrelease
Any financial support from family members

Men

76%

71%

59%

Women

62%

58%

44%

58%

Men

37%

34%

26%

Women

31%

19%

Men

54%

49%

32%

50%

Women

33%

22%

25%

Men

82%

85%

Women

74%

64%

Men

35%

29%

44%

Women

36%

44%

Men

21%

26%

28%

25%

Women

21%

Substance use
Past 30 day substance use, 8-10 months postrelease
Past 30 day substance abuse treatment services, 8-10
months postrelease
Criminal involvement
Any self-reported crime, arrest, or parole violation, 8-10
months postrelease
Reincarcerated within 12 months of release

Men

38%

45%

41%

Women

47%

53%

55%

Men

17%

20%

Women

21%

26%

Note: Data are shown when the relationship between a given health condition and outcome was statistically significant
(p≤ 0.10). Nonsignificant relationships between health conditions and outcomes are indicated as “ns.” Italics
indicate that health status influenced a given outcome differently by gender. Boldface is used to indicate subgroup
outcomes that were better than the overall sample. Note that measures of tangible family support are included
within the housing and employment domains.
Source: Returning Home original data collection in Ohio and Texas.

Health and Prisoner Reentry

Once health needs are identified, other steps can be taken to improve health outcomes after
release. A comprehensive strategy would include assessing each returning prisoner’s eligibility
for Medicaid, disability, and other forms of public assistance and beginning the application or
reinstatement process before release. All returning prisoners, including those who did not receive
health services in prison, should be educated about health care resources appropriate to their
needs in the community. Continuity of care, for those who were treated in prison, can be
facilitated by linking to appropriate providers and services in the community; individuals with
severe health conditions should have appointments with community-based providers scheduled
in advance of their release. Returning prisoners who received medications in prison should
receive a prescription or supply of medication at release, and this typically does happen; our
findings show an additional need for coordination with community-based medical providers as
well as patient education about why specific medicines were prescribed and the risks of stopping
those medicines without medical supervision.
With regard to reentry outcomes, nearly all returning prisoners with health conditions had
housing needs. They were less likely than others to have finalized their postrelease housing
arrangements when we surveyed them in the month before release, and all experienced some
degree of housing instability during the first eight to ten months after release. Similarly,
returning prisoners with health conditions also reported greater employment difficulties relative
to other returning prisoners. While the degree to which respondents experienced housing and
employment difficulties varied by the type of health condition and also by gender, these findings
suggest that all returning prisoners should receive a base level of housing and employment
assistance.
Without minimizing these challenges that all returning prisoners face, it is important to
recognize the varying degrees of difficulty faced by returning prisoners with physical health,
mental health, and substance abuse problems in order to develop targeted strategies for
improving outcomes.
Returning prisoners with physical health conditions, as a group, had the fewest distinct
challenges, yet men with physical health conditions were more likely to have been reincarcerated
within one year. Their greatest service needs after release were employment and income support.
While returning prisoners with physical health conditions initially found employment at rates
similar to others, they were less likely to have retained those jobs, and their rates of finding
employment became worse relative to healthier respondents as time elapsed. Employment
outcomes might be improved if health issues were addressed early in the postrelease period so
that health conditions would not deteriorate. Nevertheless, a subset of returning prisoners with
health problems would still need nonwork sources of financial support, and applications for
Medicaid, disability support, and other forms of public assistance should be submitted during
prison to minimize the amount of time between release and receipt of services.
Men with physical health conditions were also more likely to have been reincarcerated,
particularly for parole violations, but the reasons for this are unclear, since they were less likely
Health and Prisoner Reentry

to engage in substance use after prison and did not report elevated levels of criminal activity.
Women with physical health conditions, on the other hand, reported higher levels of criminal
behavior, yet they were not reincarcerated at higher rates. However, among women who were
reincarcerated, those with physical health conditions were more likely to have had their parole
revoked. More investigation is needed to understand the dynamics between physical health,
criminal behavior, and elevated parole revocation and reincarceration rates.
Returning prisoners with mental health conditions experienced reentry difficulties across a
range of domains and received the lowest amounts of family support, suggesting that an
intensive case management approach would be beneficial. Returning prisoners with mental
health conditions typically had co-occurring physical and substance abuse problems and reported
higher rates of homelessness, unemployment, and criminal activity relative to others. They were
also least likely to receive housing or financial support from family members, meaning that they
would require governmental or charitable assistance. Given this multitude of needs, an integrated
case management approach seems most appropriate. Interestingly, despite experiencing greater
challenges and reporting more criminal behavior, returning prisoners with mental health
conditions had not been reincarcerated at higher rates during the first postrelease year. More
research is needed to understand the criminal justice system response to persons with mental
illness.
To some extent the outcomes of returning prisoners with substance abuse problems
depended on gender. Women with substance abuse problems experienced more reentry
difficulties than men, and given their relative lack of family support systems, would require more
formal assistance than men. Housing outcomes among substance abusers, for example, differed
by gender. Men with substance abuse problems received more family support relative to other
men and were more likely to have lived with family members. By contrast, women with
substance abuse problems were less likely to receive such support and were more likely to have
been homeless. Employment difficulties were also more pronounced among women. Though
substance abusing men experienced problems keeping employment, they had at least been able to
find employment at rates similar to other men. Substance abusing women, by contrast, were less
likely to find employment and reported decreasing rates of current employment over time.
Practitioners and policymakers should recognize that men and women with substance abuse
problems require different levels of housing and employment assistance, and that substance
abusing women may be better served by a more intensive approach.
Substance abuse was clearly linked to recidivism. Both substance abusing men and women
engaged in more postrelease substance use and criminal behavior than other returning prisoners,
and they were more likely to have been reincarcerated within the first postrelease year. Men who
received substance abuse treatment services reported lower rates of postrelease use, but large
shares of returning prisoners with substance abuse problems did not receive any treatment
services while incarcerated. Increasing the amount of treatment available to substance abusing
men and women may serve to reduce aggregate rates of postrelease use. Results from this study
suggest further examination into the allocation of treatment resources, particularly in women’s

Health and Prisoner Reentry

facilities, to understand the determinants of receiving treatment services. Additional Returning
Home research is underway to gain more in-depth understanding of the roles of substance abuse
and treatment in the reentry process.
The majority of returning prisoners with health conditions received some assistance from
family members, though the level of support varied by health status and gender. Discharge
planning should consider the role of family members, including any risks posed by specific
individuals, and realistically assess the amount of support they can provide by including them in
the planning process whenever possible. Nonetheless, policymakers and practitioners should
remember that a sizeable minority—one-fifth of men and one-quarter of women—did not receive
any tangible assistance from family members. Formal assistance should be targeted to these
persons.
Regardless of the specific health condition, practitioners and policymakers should be
mindful that returning men typically experienced better outcomes than women. Furthermore, in
some instances, the relationship between health status and outcomes was different depending on
the gender of the returning prisoner, resulting in large magnitudes of difference between men and
women with similar problems. This may reflect differential treatment of men and women with
similar health conditions, but it may also be that men and women with similar health conditions
respond differently to reentry challenges. Gender differences were particularly apparent in the
area of family support, and this is important because the extent to which returning prisoners
could rely on family members often influenced their housing and financial outcomes. As a result,
returning women are more likely to need services from governmental or charitable sources.
In conclusion, we must state one important caveat. We have established the ways in which
returning prisoners with physical health, mental health, and substance abuse conditions had
reentry experiences that differed from the “average” returning prisoner. These typologies of
reentry experiences and service needs are intended to inform policymakers and practitioners
about the types of issues that returning prisoners with health conditions confront. While
informative, this cannot take the place of individual, client-centered discharge planning,
especially since so many returning prisoners reported multiple types of health conditions.
Individual returning prisoners with one type of health conditions may have needs related to other
health issues. The actual services provided to individual returning prisoners must be targeted to
their particular needs.

Health and Prisoner Reentry

Methodology

Returning Home: Understanding the Challenges of Prisoner Reentry is the Urban Institute’s
multistate, longitudinal study of the process by which people leave state prisons and resume life
in the community. Launched in 2001, this study documents the transition from prison to home in
order to identify factors that contribute to successful or unsuccessful reintegration. This research
aims to inform policy and practice by improving understanding about various reentry pathways.

The Returning Home study design focuses on the reentry process in an urban context since
prison populations are largely comprised of individuals from urban areas. In each of the four
states studied (Maryland, Illinois, Ohio, and Texas), the study selected a representative sample of
state prisoners who would soon be returning to the largest metropolitan area in that state. This
sample was surveyed at several points in time to get a longitudinal view of the reentry process.
Respondents completed a self-administered survey shortly before release and participated in oneon-one personal interviews up to three times after release. (The specific number and timing of
these interviews varied across states because of logistical and budgetary reasons.) Through these
interviews, returning prisoners shared their thoughts and experiences related to a number of
important reentry challenges, such as finding housing, gaining employment, abstaining from
substance use, and remaining crime-free. The interviews addressed factors that are hypothesized
to influence reentry success, including attitudes and expectations, physical and mental health,
substance use and treatment, family and peer relationships, and programmatic interventions. The
information collected through these interviews was supplemented with incarceration records
from each participating state’s department of corrections to measure recidivism after release.
Additionally, the study conducted focus groups in the neighborhoods to which many prisoners
return, interviewed reentry policymakers and practitioners, and reviewed relevant state laws and
policies.
This report is based on the reentry experiences of 838 men and 262 women returning from
Ohio and Texas state prisons during 2004 and 2005. In addition to direct applicability to these
areas, the findings from this study are informative about the reentry process in other major
metropolitan areas. In Ohio the study surveyed a representative sample of 424 men returning
from state prisons to Cuyahoga County (metropolitan Cleveland), and in Texas the study
surveyed a representative sample of 414 men and 262 women returning to Harris County
(metropolitan Houston). Participants first completed a self-administered survey in the month
before their release from prison. Of those surveyed in prison, 79 percent of men and 77 percent
of women completed the first postrelease interview, which occurred two to three months after
release (a mean of 68 days postrelease for men and 89 days for women). Of the original sample
members, 65 percent of men and 60 percent of women completed a second postrelease interview;
this interview occurred between eight and ten months after release (a mean of 237 days or 7.9
months postrelease for men and 291 days or 9.7 months for women). Notably, 12 percent of men

Health and Prisoner Reentry

and 19 percent of women were in prison when they participated in the second postrelease
interview; the study protocol for such interviews asked respondents to recount their experiences
in the community until the time of reincarceration. The study also obtained prison admission
records from the respective state departments of corrections on 1,073 of the 1,100 respondents in
the original sample to analyze recidivism in the first 12 months after release.
Health status was assessed through respondent self-reports at the time of the prerelease
survey and respondents were categorized as having physical health, mental health, or substance
abuse problems. Physical health status and mental health status were assessed using a checklist
of chronic physical and mental health conditions that are prevalent in correctional populations;
respondents were asked to indicate which conditions a doctor or nurse had diagnosed them with.
The study definition of substance abuse was based on the self-reported frequency of drug use or
alcohol intoxication during the six months before the current prison term; respondents who
reported preprison substance use more than once a week were categorized as substance abusers
because this frequency of use was correlated with a greater incidence of problems related to use
and more signs of addiction. Respondents with multiple types of problems (e.g., physical health
conditions and substance abuse) were included in each applicable group.
Bivariate analyses were conducted to identify gender differences in sample characteristics
and reentry outcomes. The analysis of gender differences was restricted to Texas respondents to
control for state effects because women were not sampled in the other Returning Home states.
The sample sizes for these comparisons were 414 men and 262 women (prerelease), 307 men
and 202 women (first postrelease interview), and 222 men and 156 women (second postrelease
interview). Differences presented in this report are statistically significant at p ≤ 0.10, unless
otherwise stated.
The analysis of relationships between health conditions and reentry outcomes were
conducted separately for men and for women to control for gender effects. Gender-specific
bivariate analyses were conducted to develop a profile of the reentry experiences of individuals
with each type of health condition—physical, mental, and substance abuse—and to determine
whether reentry outcomes of respondents with a given type of health condition were significantly
different from others without that condition. For example, men with physical health conditions
were compared to men without physical health conditions and women who reported preprison
substance abuse (i.e., more often than once a week) were compared to other women who did not
meet that definition of abuse. The high prevalence of respondents with any health condition—
physical, mental, or substance abuse—precluded any analysis that would compare those with any
health condition to others with none. Differences presented in this report are statistically
significant at p ≤ 0.10, unless otherwise stated.
Analyses of reentry outcomes were based on the sample size at each wave: analysis of
prerelease factors utilized the full sample of respondents and analyses of postrelease outcomes
utilized the sample size available at each postrelease time point. We examined whether the
attrition rate for respondents with physical health, mental health, and substance abuse problems

Health and Prisoner Reentry

was significantly different from others without those conditions. We found no significant
differences in eight to ten month follow-up rates. There were also no differences in two to three
month follow-up rates among women; men with physical and mental health conditions were
more likely to have participated in the two to three month postrelease interview, but the
magnitude of the difference in follow-up rates was not much. (Two to three months after release,
we conducted follow-up interviews with 82 percent of men with physical health conditions and
86 percent of men with mental health conditions, compared to 77 and 78 percent, respectively, of
men without those conditions.) Since respondents with and without health conditions had similar
rates of attrition, any bias due to sample attrition should not unduly influence the relationships
observed between health status and reentry outcomes.

Health and Prisoner Reentry

References

American Psychiatric Association. 2000. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition,
Text Revision. Washington, DC: American Psychiatric Association.
Bernstein, Jared, and Ellen Houston. 2000. Crime and Work: What We Can Learn from the Low-Wage Labor
Market. Washington, DC: Economic Policy Institute.
Centers for Disease Control and Prevention. 2003. “Prevention and Control of Infections with Hepatitis Viruses in
Correctional Settings.” Morbidity and Mortality Weekly Report 52(RR-1): 1–36.
Centers for Disease Control and Prevention. 2006. “HIV Transmission Among Male Inmates in a State Prison
System—Georgia, 1992–2005.” Morbidity and Mortality Weekly Report 55(15): 421–426.
Centers for Disease Control and Prevention. 2006. “Prevention and Control of Tuberculosis in Correctional and
Detention Facilities: Recommendations from CDC.” Morbidity and Mortality Weekly Report 55(RR-09): 1–44.
Foa, Edna B., David S. Riggs, Constance V. Dancu, and Barbara O. Rothbaum. 1993. “Reliability and Validity of a
Brief Instrument for Assessing Post-Traumatic Stress Disorder.” Journal of Traumatic Stress 6(4): 459–473.
Hammett, Theodore M., Sheryl Roberts, and Sofia Kennedy. 2001. “Health-Related Issues in Prisoner Reentry to the
Community.” Crime and Delinquency 47(3): 390–409.
Harlow, Caroline Wolf. 2003. Education and Correctional Populations. Bureau of Justice Statistics Special Report
No. NCJ 195670. Washington, DC: U.S. Department of Justice.
Holzer, Harry J., Steven Raphael, and Michael A. Stoll. 2004. “Will Employers Hire Former Offenders? Employer
Preferences, Background Checks, and Their Determinants.” In Imprisoning America: The Social Effects of Mass
Incarceration, edited by Bruce Western, Mary Patillo, and David F. Weiman. New York: The Russell Sage
Foundation.
James, Doris J., and Lauren E. Glaze. 2006. Mental Health Problems of Prison and Jail Inmates. Bureau of Justice
Statistics Special Report No. NCJ 213600. Washington, DC: U.S. Department of Justice.
Langan, Patrick A., and David J. Levin. 2002. Recidivism of Prisoners Released in 1994. Bureau of Justice Statistics
Special Report No. NCJ 193427. Washington, DC: U.S. Department of Justice.
LeBlanc, Adrian Nicole. 2003. Random Family: Love, Drugs, Trouble, and Coming of Age in the Bronx. New York:
Scribner.
Lynch, James P., and William J. Sabol. 2001. “Prisoner Reentry in Perspective.” Washington, DC: The Urban
Institute.
McLellan, A. Thomas., David C. Lewis, Charles P. O’Brien, and Herbert D. Kleber. 2000. “Drug Dependence, a
Chronic Medical Illness: Implications for Treatment, Insurance and Outcomes Evaluation.” Journal of the
American Medical Association 284(13): 1689–1695.
National Commission on Correctional Health Care. 2002. “The Health Status of Soon-to-be Released Prisoners
(Volume 1).” http://www.ncchc.org/stbr/Volume1/Health%20Status%20(vol%201).pdf.
National Institute on Drug Abuse. 2004. Cocaine Abuse and Addiction. NIDA Research Report Series No. 99-4342.
Washington, DC: National Institutes of Health.
Quello, Susan B., Kathleen T. Brady, and Susan C. Sonne. 2005. “Mood Disorders and Substance Use Disorder: A
Complex Comorbidity.” NIDA Science and Practice Perspectives, 3(1): 13–24.
Radloff, Lenore S. 1991. “The Use of the Center for Epidemiologic Studies Depression Scale in Adolescents and
Young Adults.” Journal of Youth and Adolescence 20(2): 149–166.
Sabol, William .J., Todd D. Minton, and Paige M. Harrison. 2007. Prison and Jail Inmates at Midyear 2006. Bureau
of Justice Statistics Special Report No. NCJ 217675. Washington, DC: U.S. Department of Justice.

Health and Prisoner Reentry

Travis, Jeremy. 2005. But They All Come Back: Facing the Challenges of Prisoner Reentry. Washington, DC: Urban
Institute Press.
Watson, Jamie, Amy Solomon, Nancy G. La Vigne, and Jeremy Travis. 2004. “A Portrait of Prisoner Reentry in
Texas.” Washington, DC: The Urban Institute.
Western, Bruce, and Becky Pettit. 2000. “Incarceration and Racial Inequality in Men's Employment.” Industrial and
Labor Relations Review 54(1): 3–16.

Health and Prisoner Reentry

URBAN INSTITUTE

Justice Policy Center
2100 M STREET, NW
WASHINGTON, DC 20037
www.urban.org
(202) 833-7200

The Habeas Citebook Ineffective Counsel Side